本项目每月自动从 arXiv 获取最新的论文，基于关键词进行筛选。
点击 'Watch' 按钮可以接收每月自动推送的邮件通知。
本次更新执行命令
D:\a\MyAutoPapers\MyAutoPapers\target\release\my_auto_papers.exe --keywords=
efficient RL,
partial observable markov decision process/pomdp,sparse reward reinforcement learning,
casual RL/counterfactual RL/casual reinforcement learning,
causal inference/causal discovery/counterfactual reasoning,
video super resolution,
knowledge graph/knowledge distillation/knowledge representation/knowledge transfer/knowledge embedding,
combinatorial game theory/xiangqi/chinese chess,
code llm,
speech recognition,
zero shot tracking/few shot tracking/pose tracking/pose estimation,
text to 3d/image to 3d/text to texture,
automated theorem proving/interactive theorem proving/formal verification
--exclude-keywords=multi-agent,multiagent --per-keyword-max-result=8
参数详解
- 关键词：
efficient RL,partial observable markov decision process/pomdp,sparse reward reinforcement learning,casual RL/counterfactual RL/casual reinforcement learning,causal inference/causal discovery/counterfactual reasoning,video super resolution,knowledge graph/knowledge distillation/knowledge representation/knowledge transfer/knowledge embedding,combinatorial game theory/xiangqi/chinese chess,code llm,speech recognition,zero shot tracking/few shot tracking/pose tracking/pose estimation,text to 3d/image to 3d/text to texture,automated theorem proving/interactive theorem proving/formal verification - 排除关键词：
multi-agent,multiagent - 每关键词最大结果：
8 - 目标领域：
cs,stat - 每关键词重试次数：
3
| 序号 | 标题 | 日期 | 摘要 |
|---|---|---|---|
| 1 | ThetaEvolve: Test-time Learning on Open Problems | 2025-11-28 | 展开Recent advances in large language models (LLMs) have enabled breakthroughs in mathematical discovery, exemplified by AlphaEvolve, a closed-source system that evolves programs to improve bounds on open problems. However, it relies on ensembles of frontier LLMs to achieve new bounds and is a pure inference system that models cannot internalize the evolving strategies. We introduce ThetaEvolve, an open-source framework that simplifies and extends AlphaEvolve to efficiently scale both in-context learning and Reinforcement Learning (RL) at test time, allowing models to continually learn from their experiences in improving open optimization problems. ThetaEvolve features a single LLM, a large program database for enhanced exploration, batch sampling for higher throughput, lazy penalties to discourage stagnant outputs, and optional reward shaping for stable training signals, etc. ThetaEvolve is the first evolving framework that enable a small open-source model, like DeepSeek-R1-0528-Qwen3-8B, to achieve new best-known bounds on open problems (circle packing and first auto-correlation inequality) mentioned in AlphaEvolve. Besides, across two models and four open tasks, we find that ThetaEvolve with RL at test-time consistently outperforms inference-only baselines, and the model indeed learns evolving capabilities, as the RL-trained checkpoints demonstrate faster progress and better final performance on both trained target task and other unseen tasks. We release our code publicly: https://github.com/ypwang61/ThetaEvolve |
| 2 | Improving Stochastic Action-Constrained Reinforcement Learning via Truncated Distributions | 2025-11-27 | 展开In reinforcement learning (RL), it is often advantageous to consider additional constraints on the action space to ensure safety or action relevance. Existing work on such action-constrained RL faces challenges regarding effective policy updates, computational efficiency, and predictable runtime. Recent work proposes to use truncated normal distributions for stochastic policy gradient methods. However, the computation of key characteristics, such as the entropy, log-probability, and their gradients, becomes intractable under complex constraints. Hence, prior work approximates these using the non-truncated distributions, which severely degrades performance. We argue that accurate estimation of these characteristics is crucial in the action-constrained RL setting, and propose efficient numerical approximations for them. We also provide an efficient sampling strategy for truncated policy distributions and validate our approach on three benchmark environments, which demonstrate significant performance improvements when using accurate estimations. |
| 3 | TinyLLM: Evaluation and Optimization of Small Language Models for Agentic Tasks on Edge Devices | 2025-11-27 | 展开This paper investigates the effectiveness of small language models (SLMs) for agentic tasks (function/tool/API calling) with a focus on running agents on edge devices without reliance on cloud infrastructure. We evaluate SLMs using the Berkeley Function Calling Leaderboard (BFCL) framework and describe parameter-driven optimization strategies that include supervised fine-tuning (SFT), parameter-efficient fine-tuning (PEFT), reinforcement learning (RL)-based optimization, preference alignment via Direct Preference Optimization (DPO), and hybrid methods. We report results for models including TinyAgent, TinyLlama, Qwen, and xLAM across BFCL categories (simple, multiple, parallel, parallel-multiple, and relevance detection), both in live and non-live settings, and in multi-turn evaluations. We additionally detail a DPO training pipeline constructed from AgentBank data (e.g., ALFRED), including our conversion of SFT data to chosen-rejected pairs using TinyLlama responses as rejected outputs and manual validation. Our results demonstrate clear accuracy differences across model scales where medium-sized models (1-3B parameters) significantly outperform ultra-compact models (<1B parameters), achieving up to 65.74% overall accuracy, and 55.62% multi-turn accuracy with hybrid optimization. This study highlights the importance of hybrid optimization strategies that enable small language models to deliver accurate, efficient, and stable agentic AI on edge devices, making privacy-preserving, low-latency autonomous agents practical beyond the cloud. |
| 4 | An energy-efficient spiking neural network with continuous learning for self-adaptive brain-machine interface | 2025-11-27 | 展开The number of simultaneously recorded neurons follows an exponentially increasing trend in implantable brain-machine interfaces (iBMIs). Integrating the neural decoder in the implant is an effective data compression method for future wireless iBMIs. However, the non-stationarity of the system makes the performance of the decoder unreliable. To avoid frequent retraining of the decoder and to ensure the safety and comfort of the iBMI user, continuous learning is essential for real-life applications. Since Deep Spiking Neural Networks (DSNNs) are being recognized as a promising approach for developing resource-efficient neural decoder, we propose continuous learning approaches with Reinforcement Learning (RL) algorithms adapted for DSNNs. Banditron and AGREL are chosen as the two candidate RL algorithms since they can be trained with limited computational resources, effectively addressing the non-stationary problem and fitting the energy constraints of implantable devices. To assess the effectiveness of the proposed methods, we conducted both open-loop and closed-loop experiments. The accuracy of open-loop experiments conducted with DSNN Banditron and DSNN AGREL remains stable over extended periods. Meanwhile, the time-to-target in the closed-loop experiment with perturbations, DSNN Banditron performed comparably to that of DSNN AGREL while achieving reductions of 98% in memory access usage and 99% in the requirements for multiply- and-accumulate (MAC) operations during training. Compared to previous continuous learning SNN decoders, DSNN Banditron requires 98% less computes making it a prime candidate for future wireless iBMI systems. |
| 5 | Representative Action Selection for Large Action Space: From Bandits to MDPs | 2025-11-27 | 展开We study the problem of selecting a small, representative action subset from an extremely large action space shared across a family of reinforcement learning (RL) environments -- a fundamental challenge in applications like inventory management and recommendation systems, where direct learning over the entire space is intractable. Our goal is to identify a fixed subset of actions that, for every environment in the family, contains a near-optimal action, thereby enabling efficient learning without exhaustively evaluating all actions. This work extends our prior results for meta-bandits to the more general setting of Markov Decision Processes (MDPs). We prove that our existing algorithm achieves performance comparable to using the full action space. This theoretical guarantee is established under a relaxed, non-centered sub-Gaussian process model, which accommodates greater environmental heterogeneity. Consequently, our approach provides a computationally and sample-efficient solution for large-scale combinatorial decision-making under uncertainty. |
| 6 | OpenMMReasoner: Pushing the Frontiers for Multimodal Reasoning with an Open and General Recipe | 2025-11-20 | 展开Recent advancements in large reasoning models have fueled growing interest in extending such capabilities to multimodal domains. However, despite notable progress in visual reasoning, the lack of transparent and reproducible data curation and training strategies remains a major barrier to scalable research. In this work, we introduce OpenMMReasoner, a fully transparent two-stage recipe for multimodal reasoning spanning supervised fine-tuning (SFT) and reinforcement learning (RL). In the SFT stage, we construct an 874K-sample cold-start dataset with rigorous step-by-step validation, providing a strong foundation for reasoning capabilities. The subsequent RL stage leverages a 74K-sample dataset across diverse domains to further sharpen and stabilize these abilities, resulting in a more robust and efficient learning process. Extensive evaluations demonstrate that our training recipe not only surpasses strong baselines but also highlights the critical role of data quality and training design in shaping multimodal reasoning performance. Notably, our method achieves a 11.6% improvement over the Qwen2.5-VL-7B-Instruct baseline across nine multimodal reasoning benchmarks, establishing a solid empirical foundation for future large-scale multimodal reasoning research. We open-sourced all our codes, pipeline, and data at https://github.com/EvolvingLMMs-Lab/OpenMMReasoner. |
| 7 | Asymmetric REINFORCE for off-Policy Reinforcement Learning: Balancing positive and negative rewards | 2025-06-25 | 展开Reinforcement learning (RL) is increasingly used to align large language models (LLMs). Off-policy methods offer greater implementation simplicity and data efficiency than on-policy techniques, but often result in suboptimal performance. In this work, we study the intermediate range of algorithms between off-policy RL and supervised fine-tuning by analyzing a simple off-policy REINFORCE algorithm, where the advantage is defined as |
| 序号 | 标题 | 日期 | 摘要 |
|---|---|---|---|
| 1 | From CAD to POMDP: Probabilistic Planning for Robotic Disassembly of End-of-Life Products | 2025-11-28 | 展开To support the circular economy, robotic systems must not only assemble new products but also disassemble end-of-life (EOL) ones for reuse, recycling, or safe disposal. Existing approaches to disassembly sequence planning often assume deterministic and fully observable product models, yet real EOL products frequently deviate from their initial designs due to wear, corrosion, or undocumented repairs. We argue that disassembly should therefore be formulated as a Partially Observable Markov Decision Process (POMDP), which naturally captures uncertainty about the product's internal state. We present a mathematical formulation of disassembly as a POMDP, in which hidden variables represent uncertain structural or physical properties. Building on this formulation, we propose a task and motion planning framework that automatically derives specific POMDP models from CAD data, robot capabilities, and inspection results. To obtain tractable policies, we approximate this formulation with a reinforcement-learning approach that operates on stochastic action outcomes informed by inspection priors, while a Bayesian filter continuously maintains beliefs over latent EOL conditions during execution. Using three products on two robotic systems, we demonstrate that this probabilistic planning framework outperforms deterministic baselines in terms of average disassembly time and variance, generalizes across different robot setups, and successfully adapts to deviations from the CAD model, such as missing or stuck parts. |
| 2 | Does Self-Evaluation Enable Wireheading in Language Models? | 2025-11-28 | 展开Self-evaluation is increasingly central to language model training, from constitutional AI to self-refinement. We investigate whether coupling self-evaluation to reward signals creates incentives for wireheading, where agents manipulate reward measurements rather than improving task performance. We formalize conditions under which reward-channel control strictly dominates task-focused behavior in POMDPs and test these predictions empirically. Across two models and three tasks, we find that models whose self-grades determine rewards exhibit substantial grade inflation without corresponding accuracy gains, particularly on ambiguous tasks like summarization. Models that self-evaluate but do not control rewards show no such inflation. Our results demonstrate that self-evaluation is safe when decoupled from learning signals but dangerous when coupled, with clear implications for agentic system design. |
| 3 | Entropy is all you need for Inter-Seed Cross-Play in Hanabi | 2025-11-27 | 展开We find that in Hanabi, one of the most complex and popular benchmarks for zero-shot coordination and ad-hoc teamplay, a standard implementation of independent PPO with a slightly higher entropy coefficient 0.05 instead of the typically used 0.01, achieves a new state-of-the-art in cross-play between different seeds, beating by a significant margin all previous specialized algorithms, which were specifically designed for this setting. We provide an intuition for why sufficiently high entropy regularization ensures that different random seed produce joint policies which are mutually compatible. We also empirically find that a high |
| 4 | ENACT: Evaluating Embodied Cognition with World Modeling of Egocentric Interaction | 2025-11-26 | 展开Embodied cognition argues that intelligence arises from sensorimotor interaction rather than passive observation. It raises an intriguing question: do modern vision-language models (VLMs), trained largely in a disembodied manner, exhibit signs of embodied cognition? We introduce ENACT, a benchmark that casts evaluation of embodied cognition as world modeling from egocentric interaction in a visual question answering (VQA) format. Framed as a partially observable Markov decision process (POMDP) whose actions are scene graph changes, ENACT comprises two complementary sequence reordering tasks: forward world modeling (reorder shuffled observations given actions) and inverse world modeling (reorder shuffled actions given observations). While conceptually simple, solving these tasks implicitly demands capabilities central to embodied cognition-affordance recognition, action-effect reasoning, embodied awareness, and interactive, long-horizon memory from partially observable egocentric input, while avoiding low-level image synthesis that could confound the evaluation. We provide a scalable pipeline that synthesizes QA pairs from robotics simulation (BEHAVIOR) and evaluates models on 8,972 QA pairs spanning long-horizon home-scale activities. Experiments reveal a performance gap between frontier VLMs and humans that widens with interaction horizon. Models consistently perform better on the inverse task than the forward one and exhibit anthropocentric biases, including a preference for right-handed actions and degradation when camera intrinsics or viewpoints deviate from human vision. Website at https://enact-embodied-cognition.github.io/. |
| 5 | POMDP-Based Routing for DTNs with Partial Knowledge and Dependent Failures | 2025-11-25 | 展开Routing in Delay-Tolerant Networks (DTNs) is inherently challenging due to sparse connectivity, long delays, and frequent disruptions. While Markov Decision Processes (MDPs) have been used to model uncertainty, they assume full state observability - an assumption that breaks down in partitioned DTNs, where each node operates with inherently partial knowledge of the network state. In this work, we investigate the role of Partially Observable Markov Decision Processes (POMDPs) for DTN routing under uncertainty. We introduce and evaluate a novel model: Dependent Node Failures (DNF), which captures correlated node failures via repairable node states modeled as Continuous-Time Markov Chains (CTMCs). We implement the model using JuliaPOMDP and integrate it with DTN simulations via DtnSim. Our evaluation demonstrates that POMDP-based routing yields improved delivery ratios and delay performance under uncertain conditions while maintaining scalability. These results highlight the potential of POMDPs as a principled foundation for decision-making in future DTN deployments. |
| 6 | AVA-VLA: Improving Vision-Language-Action models with Active Visual Attention | 2025-11-24 | 展开Vision-Language-Action (VLA) models have demonstrated remarkable capabilities in embodied AI tasks. However, existing VLA models, often built upon Vision-Language Models (VLMs), typically process dense visual inputs independently at each timestep. This approach implicitly models the task as a Markov Decision Process (MDP). However, this history-agnostic design is suboptimal for effective visual token processing in dynamic sequential decision-making, as it fails to leverage the context of history. To address this limitation, we reformulate the problem from a Partially Observable Markov Decision Process (POMDP) perspective and propose a novel framework named AVA-VLA. Inspired by the POMDP that the action generation should be conditioned on the belief state. AVA-VLA introduces Active Visual Attention (AVA) to dynamically modulate visual processing. It achieves this by leveraging the recurrent state, which is a neural approximation of the agent's belief state derived from the previous decision step. Specifically, the AVA module uses the recurrent state to compute the soft weights to actively process task-relevant visual tokens based on its historical context. Comprehensive evaluations demonstrate that AVA-VLA achieves state-of-the-art performance across popular robotic benchmarks, including LIBERO and CALVIN. Furthermore, real-world deployments on a dual-arm robot platform validate the framework's practical applicability and robust sim-to-real transferability. |
| 7 | Delay-Optimal Transmission Scheduling Policies for Time-Correlated Fading Channels | 2025-11-22 | 展开Millimeter-wave (mmWave) networks have the potential to support high throughput and low-latency requirements of 5G-and-beyond communication standards. But transmissions in this band are highly vulnerable to attenuation and blockages from humans, buildings, and foliage, which increase end-to-end packet delays. This work designs dynamic scheduling policies that minimize end-to-end packet delays while keeping packet transmission costs low. Specifically, we consider a mmWave network that consists of a transmitter that transmits data packets over an unreliable communication channel modeled as a Gilbert-Elliott channel.The transmitter operates under an ACK/NACK feedback model and does not observe the channel state unless it attempts a transmission. The objective is to minimize a weighted average cost consisting of end-to-end packet delays and packet transmission costs. We pose this dynamic optimization problem as a partially observable Markov decision process (POMDP). To the best of our knowledge, this is the first POMDP formulation for mmWave network with partial channel state information that considers delay minimization. We show that the POMDP admits a solution that has a threshold structure, i.e., for each queue length, the belief (the conditional probability that the channel is in a good state) is partitioned into intervals, and the transmitter sends j packets when the belief lies in the j-th interval. We then consider the case when the system parameters such as the packet arrival rate, and the transition probabilities of the channel are not known, and leverage these structural results in order to use the actor-critic algorithm to efficiently search for a policy that is locally optimal. |
| 8 | Vectorized Online POMDP Planning | 2025-10-31 | 展开Planning under partial observability is an essential capability of autonomous robots. The Partially Observable Markov Decision Process (POMDP) provides a powerful framework for planning under partial observability problems, capturing the stochastic effects of actions and the limited information available through noisy observations. POMDP solving could benefit tremendously from massive parallelization of today's hardware, but parallelizing POMDP solvers has been challenging. They rely on interleaving numerical optimization over actions with the estimation of their values, which creates dependencies and synchronization bottlenecks between parallel processes that can quickly offset the benefits of parallelization. In this paper, we propose Vectorized Online POMDP Planner (VOPP), a novel parallel online solver that leverages a recent POMDP formulation that analytically solves part of the optimization component, leaving only the estimation of expectations for numerical computation. VOPP represents all data structures related to planning as a collection of tensors and implements all planning steps as fully vectorized computations over this representation. The result is a massively parallel solver with no dependencies and synchronization bottlenecks between parallel computations. Experimental results indicate that VOPP is at least 20X more efficient in computing near-optimal solutions compared to an existing state-of-the-art parallel online solver. |
| 9 | Observer-Aware Probabilistic Planning Under Partial Observability | 2025-02-14 | 展开In this article, we are interested in planning problems where the agent is aware of the presence of an observer, and where this observer is in a partial observability situation. The agent has to choose its strategy so as to optimize the information transmitted by observations. Building on observer-aware Markov decision processes (OAMDPs), we propose a framework to handle this type of problems and thus formalize properties such as legibility, explicability and predictability. This extension of OAMDPs to partial observability can not only handle more realistic problems, but also permits considering dynamic hidden variables of interest. These dynamic target variables allow, for instance, working with predictability, or with legibility problems where the goal might change during execution. We discuss theoretical properties of PO-OAMDPs and, experimenting with benchmark problems, we analyze HSVI's convergence behavior with dedicated initializations and study the resulting strategies. |
| 10 | Uniform Value and Decidability in Ergodic Blind Stochastic Games | 2024-05-21 | 展开We study a class of two-player zero-sum stochastic games known as \textit{blind stochastic games}, where players neither observe the state nor receive any information about it during the game. A central concept for analyzing long-duration stochastic games is the \textit{uniform value}. A game has a uniform value |
| 序号 | 标题 | 日期 | 摘要 |
|---|---|---|---|
| 1 | Video-CoM: Interactive Video Reasoning via Chain of Manipulations | 2025-11-28 | 展开Recent multimodal large language models (MLLMs) have advanced video understanding, yet most still "think about videos" ie once a video is encoded, reasoning unfolds entirely in text, treating visual input as a static context. This passive paradigm creates a semantic bottleneck: models cannot rewatch, refocus, or verify evidence, leading to shallow visual reasoning on tasks requiring fine grained spatio temporal understanding. In this work, we introduce Interactive Video Reasoning, a new paradigm that transforms video into an active cognitive workspace, enabling models to "think with videos". Our model, Video CoM, reasons through a Chain of Manipulations (CoM), performing iterative visual actions to gather and refine evidence. To support this behavior, we construct Video CoM Instruct, an 18K instruction tuning dataset curated for multi step manipulation reasoning. Beyond supervised learning, we further optimize the manipulation policy via reinforcement learning with reasoning aware Group Relative Policy Optimization (GRPO). Unlike prior work that relies solely on sparse answer rewards, our method introduces step level reasoning rewards, guiding the model toward grounded and consistent reasoning. Video CoM achieves strong results across nine video reasoning benchmarks, improving average performance by 3.6 percent over recent state of the art models, while training on only 25K SFT and 3K GRPO video samples, significantly fewer than comparable large scale models. Ablation studies demonstrate that reasoning aware rewards improve both accuracy and interpretability. Code: https://github.com/mbzuai-oryx/Video-CoM |
| 2 | Hybrid-AIRL: Enhancing Inverse Reinforcement Learning with Supervised Expert Guidance | 2025-11-26 | 展开Adversarial Inverse Reinforcement Learning (AIRL) has shown promise in addressing the sparse reward problem in reinforcement learning (RL) by inferring dense reward functions from expert demonstrations. However, its performance in highly complex, imperfect-information settings remains largely unexplored. To explore this gap, we evaluate AIRL in the context of Heads-Up Limit Hold'em (HULHE) poker, a domain characterized by sparse, delayed rewards and significant uncertainty. In this setting, we find that AIRL struggles to infer a sufficiently informative reward function. To overcome this limitation, we contribute Hybrid-AIRL (H-AIRL), an extension that enhances reward inference and policy learning by incorporating a supervised loss derived from expert data and a stochastic regularization mechanism. We evaluate H-AIRL on a carefully selected set of Gymnasium benchmarks and the HULHE poker setting. Additionally, we analyze the learned reward function through visualization to gain deeper insights into the learning process. Our experimental results show that H-AIRL achieves higher sample efficiency and more stable learning compared to AIRL. This highlights the benefits of incorporating supervised signals into inverse RL and establishes H-AIRL as a promising framework for tackling challenging, real-world settings. |
| 3 | Aligning LLMs with Biomedical Knowledge using Balanced Fine-Tuning | 2025-11-26 | 展开Effective post-training is essential to align Large Language Models (LLMs) with specialized biomedical knowledge to accelerate life science research. However, current approaches face significant limitations. First, biomedical reasoning involves intricate mechanisms often represented by sparse textual data. Standard Supervised Fine-Tuning (SFT) tends to overfit to surface-level instruction patterns without effectively internalizing this fragmented scientific knowledge. Second, Reinforcement Learning (RL) is impractical for this domain, as defining meaningful rewards often necessitates prohibitive experimental validation (e.g., wet-lab verification of drug responses), rendering real-time feedback unfeasible. We propose Balanced Fine-Tuning (BFT), an efficient post-training method designed to learn complex reasoning from sparse data without external reward signals. BFT operates through a two-layer weighting mechanism: 1. At the token level, it scales loss via prediction probabilities to stabilize gradients and prevent overfitting; 2. At the sample level, it uses "minimum group confidence" to adaptively enhance the learning of hard samples. Experiments demonstrate that BFT significantly outperforms SFT. In medical tasks, it enables LLMs to acquire knowledge that SFT misses. In biological tasks, BFT-based LLMs surpass GeneAgent (an accurate agent for biology analysis) in biological process reasoning. Moreover, the text embeddings generated by BFT can be directly applied to downstream tasks, such as gene interaction and single-cell perturbation response prediction. These results indicate that BFT facilitates broad applications of LLMs in biomedical research. |
| 4 | ReEXplore: Improving MLLMs for Embodied Exploration with Contextualized Retrospective Experience Replay | 2025-11-24 | 展开Embodied exploration is a target-driven process that requires embodied agents to possess fine-grained perception and knowledge-enhanced decision making. While recent attempts leverage MLLMs for exploration due to their strong perceptual and reasoning abilities, we find that MLLM-based embodied agents remain suboptimal in exploring new environments: (i) they rely on profound but stale pre-trained knowledge, (ii) training-based approaches such as imitation learning or reinforcement learning are expensive for long-horizon tasks with sparse outcome rewards, and (iii) frontier-based exploration yields a large, visually nuanced action space that is difficult for MLLMs to make reliable decisions. We address these challenges with ReEXplore, a training-free framework that performs retrospective experience replay to inject distilled, abstract experience at inference time, and hierarchical frontier selection to decompose frontier ranking into coarse-to-fine decisions. Our approach enables robust, traceable, and efficient exploration. Across multiple embodied exploration benchmarks, ReEXplore yields great improvements over strong MLLM baselines, up to 3x higher performance in both success rate and in navigation efficiency under open-source backbones. |
| 5 | Learning to Compress Graphs via Dual Agents for Consistent Topological Robustness Evaluation | 2025-11-24 | 展开As graph-structured data grow increasingly large, evaluating their robustness under adversarial attacks becomes computationally expensive and difficult to scale. To address this challenge, we propose to compress graphs into compact representations that preserve both topological structure and robustness profile, enabling efficient and reliable evaluation. We propose Cutter, a dual-agent reinforcement learning framework composed of a Vital Detection Agent (VDA) and a Redundancy Detection Agent (RDA), which collaboratively identify structurally vital and redundant nodes for guided compression. Cutter incorporates three key strategies to enhance learning efficiency and compression quality: trajectory-level reward shaping to transform sparse trajectory returns into dense, policy-equivalent learning signals; prototype-based shaping to guide decisions using behavioral patterns from both high- and low-return trajectories; and cross-agent imitation to enable safer and more transferable exploration. Experiments on multiple real-world graphs demonstrate that Cutter generates compressed graphs that retain essential static topological properties and exhibit robustness degradation trends highly consistent with the original graphs under various attack scenarios, thereby significantly improving evaluation efficiency without compromising assessment fidelity. |
| 6 | Accelerating Reinforcement Learning via Error-Related Human Brain Signals | 2025-11-24 | 展开In this work, we investigate how implicit neural feed back can accelerate reinforcement learning in complex robotic manipulation settings. While prior electroencephalogram (EEG) guided reinforcement learning studies have primarily focused on navigation or low-dimensional locomotion tasks, we aim to understand whether such neural evaluative signals can improve policy learning in high-dimensional manipulation tasks involving obstacles and precise end-effector control. We integrate error related potentials decoded from offline-trained EEG classifiers into reward shaping and systematically evaluate the impact of human-feedback weighting. Experiments on a 7-DoF manipulator in an obstacle-rich reaching environment show that neural feedback accelerates reinforcement learning and, depending on the human-feedback weighting, can yield task success rates that at times exceed those of sparse-reward baselines. Moreover, when applying the best-performing feedback weighting across all sub jects, we observe consistent acceleration of reinforcement learning relative to the sparse-reward setting. Furthermore, leave-one subject-out evaluations confirm that the proposed framework remains robust despite the intrinsic inter-individual variability in EEG decodability. Our findings demonstrate that EEG-based reinforcement learning can scale beyond locomotion tasks and provide a viable pathway for human-aligned manipulation skill acquisition. |
| 7 | Learn the Ropes, Then Trust the Wins: Self-imitation with Progressive Exploration for Agentic Reinforcement Learning | 2025-09-26 | 展开Reinforcement learning (RL) is the dominant paradigm for sharpening strategic tool use capabilities of LLMs on long-horizon, sparsely-rewarded agent tasks, yet it faces a fundamental challenge of exploration-exploitation trade-off. Existing studies stimulate exploration through the lens of policy entropy, but such mechanical entropy maximization is prone to RL instability due to the multi-turn distribution shifting. In this paper, we target the progressive exploration-exploitation balance under the guidance of the agent's own experiences without succumbing to either entropy collapsing or runaway divergence. We propose SPEAR, a self-imitation learning (SIL) recipe for training agentic LLMs. It extends the vanilla SIL, where a replay buffer stores good experience for off-policy update, by gradually steering the policy entropy across stages. Specifically, the proposed curriculum scheduling harmonizes intrinsic reward shaping and self-imitation to 1) expedite exploration via frequent tool interactions at the beginning, and 2) strengthen exploitation of successful tactics upon convergence towards familiarity with the environment. We also combine bag-of-tricks of industrial RL optimizations for a strong baseline Dr.BoT to demonstrate our effectiveness. In ALFWorld and WebShop, SPEAR increases the success rates of GRPO/GiGPO/Dr.BoT by up to 16.1%/5.1%/8.6% and 20.7%/11.8%/13.9%, respectively. In AIME24 and AIME25, SPEAR boosts Dr.BoT by up to 3.8% and 6.1%, respectively. Such gains incur only 10%-25% extra theoretical complexity and negligible runtime overhead in practice, demonstrating the plug-and-play scalability of SPEAR. |
| 8 | Interpretable Reward Model via Sparse Autoencoder | 2025-08-12 | 展开Large language models (LLMs) have been widely deployed across numerous fields. Reinforcement Learning from Human Feedback (RLHF) leverages reward models (RMs) as proxies for human preferences to align LLM behaviors with human values, making the accuracy, reliability, and interpretability of RMs critical for effective alignment. However, traditional RMs lack interpretability, offer limited insight into the reasoning behind reward assignments, and are inflexible toward user preference shifts. While recent multidimensional RMs aim for improved interpretability, they often fail to provide feature-level attribution and require costly annotations. To overcome these limitations, we introduce the Sparse Autoencoder-enhanced Reward Model (SARM), a novel architecture that integrates a pretrained Sparse Autoencoder (SAE) into a reward model. SARM maps the hidden activations of LLM-based RM into an interpretable, sparse, and monosemantic feature space, from which a scalar head aggregates feature activations to produce transparent and conceptually meaningful reward scores. Empirical evaluations demonstrate that SARM facilitates direct feature-level attribution of reward assignments, allows dynamic adjustment to preference shifts, and achieves superior alignment performance compared to conventional reward models. Our code is available at https://github.com/schrieffer-z/sarm. |
| 序号 | 标题 | 日期 | 摘要 |
|---|---|---|---|
| 1 | AD-R1: Closed-Loop Reinforcement Learning for End-to-End Autonomous Driving with Impartial World Models | 2025-11-25 | 展开End-to-end models for autonomous driving hold the promise of learning complex behaviors directly from sensor data, but face critical challenges in safety and handling long-tail events. Reinforcement Learning (RL) offers a promising path to overcome these limitations, yet its success in autonomous driving has been elusive. We identify a fundamental flaw hindering this progress: a deep seated optimistic bias in the world models used for RL. To address this, we introduce a framework for post-training policy refinement built around an Impartial World Model. Our primary contribution is to teach this model to be honest about danger. We achieve this with a novel data synthesis pipeline, Counterfactual Synthesis, which systematically generates a rich curriculum of plausible collisions and off-road events. This transforms the model from a passive scene completer into a veridical forecaster that remains faithful to the causal link between actions and outcomes. We then integrate this Impartial World Model into our closed-loop RL framework, where it serves as an internal critic. During refinement, the agent queries the critic to ``dream" of the outcomes for candidate actions. We demonstrate through extensive experiments, including on a new Risk Foreseeing Benchmark, that our model significantly outperforms baselines in predicting failures. Consequently, when used as a critic, it enables a substantial reduction in safety violations in challenging simulations, proving that teaching a model to dream of danger is a critical step towards building truly safe and intelligent autonomous agents. |
| 2 | On a Reinforcement Learning Methodology for Epidemic Control, with application to COVID-19 | 2025-11-22 | 展开This paper presents a real time, data driven decision support framework for epidemic control. We combine a compartmental epidemic model with sequential Bayesian inference and reinforcement learning (RL) controllers that adaptively choose intervention levels to balance disease burden, such as intensive care unit (ICU) load, against socio economic costs. We construct a context specific cost function using empirical experiments and expert feedback. We study two RL policies: an ICU threshold rule computed via Monte Carlo grid search, and a policy based on a posterior averaged Q learning agent. We validate the framework by fitting the epidemic model to publicly available ICU occupancy data from the COVID 19 pandemic in England and then generating counterfactual roll out scenarios under each RL controller, which allows us to compare the RL policies to the historical government strategy. Over a 300 day period and for a range of cost parameters, both controllers substantially reduce ICU burden relative to the observed interventions, illustrating how Bayesian sequential learning combined with RL can support the design of epidemic control policies. |
| 3 | Do AI Voices Learn Social Nuances? A Case of Politeness and Speech Rate | 2025-11-12 | 展开Voice-based artificial intelligence is increasingly expected to adhere to human social conventions, but can it learn implicit cues that are not explicitly programmed? This study investigates whether state-of-the-art text-to-speech systems have internalized the human tendency to reduce speech rate to convey politeness - a non-obvious prosodic marker. We prompted 22 synthetic voices from two leading AI platforms (AI Studio and OpenAI) to read a fixed script under both "polite and formal" and "casual and informal" conditions and measured the resulting speech duration. Across both AI platforms, the polite prompt produced slower speech than the casual prompt with very large effect sizes, an effect that was statistically significant for all of AI Studio's voices and for a large majority of OpenAI's voices. These results demonstrate that AI can implicitly learn and replicate psychological nuances of human communication, highlighting its emerging role as a social actor capable of reinforcing human social norms. |
| 4 | Safe, Efficient, and Robust Reinforcement Learning for Ranking and Diffusion Models | 2025-10-17 | 展开This dissertation investigates how reinforcement learning (RL) methods can be designed to be safe, sample-efficient, and robust. Framed through the unifying perspective of contextual-bandit RL, the work addresses two major application domains - ranking and recommendation, and text-to-image diffusion models. The first part of the thesis develops theory and algorithms for safe deployment in ranking systems. An exposure-based generalisation bound is derived, leading to a counterfactual risk-minimisation objective whose solution is guaranteed not to underperform the logging policy, even with sparse feedback. This guarantee is extended to doubly robust estimators, enabling safety even under adversarial or misspecified user models and offering practitioners explicit control over permissible utility loss. The second part turns to single-action bandits, where various off-policy estimators are unified within a baseline-correction framework. A closed-form optimal baseline is proposed and shown to minimise both evaluation and policy-gradient variance, thereby improving off-policy learning reliability. The final part examines the trade-offs between efficiency and effectiveness in generative RL. A systematic study of PPO and REINFORCE motivates the Leave-One-Out PPO (LOOP) algorithm, which combines multiple diffusion trajectories with a REINFORCE-style baseline inside PPO's clipped objective. LOOP achieves PPO-level sample efficiency while producing generations that align more faithfully with textual attributes. |
| 5 | Rethinking RL Evaluation: Can Benchmarks Truly Reveal Failures of RL Methods? | 2025-10-12 | 展开Current benchmarks are inadequate for evaluating progress in reinforcement learning (RL) for large language models (LLMs).Despite recent benchmark gains reported for RL, we find that training on these benchmarks' training sets achieves nearly the same performance as training directly on the test sets, suggesting that the benchmarks cannot reliably separate further progress.To study this phenomenon, we introduce a diagnostic suite and the Oracle Performance Gap (OPG) metric that quantifies the performance difference between training on the train split versus the test split of a benchmark. We further analyze this phenomenon with stress tests and find that, despite strong benchmark scores, existing RL methods struggle to generalize across distribution shifts, varying levels of difficulty, and counterfactual scenarios: shortcomings that current benchmarks fail to reveal.We conclude that current benchmarks are insufficient for evaluating generalization and propose three core principles for designing more faithful benchmarks: sufficient difficulty, balanced evaluation, and distributional robustness. |
| 6 | Data-driven simulator of multi-animal behavior with unknown dynamics via offline and online reinforcement learning | 2025-10-12 | 展开Simulators of animal movements play a valuable role in studying behavior. Advances in imitation learning for robotics have expanded possibilities for reproducing human and animal movements. A key challenge for realistic multi-animal simulation in biology is bridging the gap between unknown real-world transition models and their simulated counterparts. Because locomotion dynamics are seldom known, relying solely on mathematical models is insufficient; constructing a simulator that both reproduces real trajectories and supports reward-driven optimization remains an open problem. We introduce a data-driven simulator for multi-animal behavior based on deep reinforcement learning and counterfactual simulation. We address the ill-posed nature of the problem caused by high degrees of freedom in locomotion by estimating movement variables of an incomplete transition model as actions within an RL framework. We also employ a distance-based pseudo-reward to align and compare states between cyber and physical spaces. Validated on artificial agents, flies, newts, and silkmoth, our approach achieves higher reproducibility of species-specific behaviors and improved reward acquisition compared with standard imitation and RL methods. Moreover, it enables counterfactual behavior prediction in novel experimental settings and supports multi-individual modeling for flexible what-if trajectory generation, suggesting its potential to simulate and elucidate complex multi-animal behaviors. |
| 7 | PuzzleJAX: A Benchmark for Reasoning and Learning | 2025-08-22 | 展开We introduce PuzzleJAX, a GPU-accelerated puzzle game engine and description language designed to support rapid benchmarking of tree search, reinforcement learning, and LLM reasoning abilities. Unlike existing GPU-accelerated learning environments that provide hard-coded implementations of fixed sets of games, PuzzleJAX allows dynamic compilation of any game expressible in its domain-specific language (DSL). This DSL follows PuzzleScript, which is a popular and accessible online game engine for designing puzzle games. In this paper, we validate in PuzzleJAX several hundred of the thousands of games designed in PuzzleScript by both professional designers and casual creators since its release in 2013, thereby demonstrating PuzzleJAX's coverage of an expansive, expressive, and human-relevant space of tasks. By analyzing the performance of search, learning, and language models on these games, we show that PuzzleJAX can naturally express tasks that are both simple and intuitive to understand, yet often deeply challenging to master, requiring a combination of control, planning, and high-level insight. |
| 8 | Pushdown Reward Machines for Reinforcement Learning | 2025-08-09 | 展开Reward machines (RMs) are automata structures that encode (non-Markovian) reward functions for reinforcement learning (RL). RMs can reward any behaviour representable in regular languages and, when paired with RL algorithms that exploit RM structure, have been shown to significantly improve sample efficiency in many domains. In this work, we present pushdown reward machines (pdRMs), an extension of reward machines based on deterministic pushdown automata. pdRMs can recognise and reward temporally extended behaviours representable in deterministic context-free languages, making them more expressive than reward machines. We introduce two variants of pdRM-based policies, one which has access to the entire stack of the pdRM, and one which can only access the top |
| 9 | AgentStealth: Reinforcing Large Language Model for Anonymizing User-generated Text | 2025-06-26 | 展开In today's digital world, casual user-generated content often contains subtle cues that may inadvertently expose sensitive personal attributes. Such risks underscore the growing importance of effective text anonymization to safeguard individual privacy. However, existing methods either rely on rigid replacements that damage utility or cloud-based LLMs that are costly and pose privacy risks. To address these issues, we explore the use of locally deployed smaller-scale language models (SLMs) for anonymization. Yet training effective SLMs remains challenging due to limited high-quality supervision. To address the challenge, we propose AgentStealth, a self-reinforcing LLM anonymization framework.First, we introduce an adversarial anonymization workflow enhanced by In-context Contrastive Learning and Adaptive Utility-Aware Control. Second, we perform supervised adaptation of SLMs using high-quality data collected from the workflow, which includes both anonymization and attack signals. Finally, we apply online reinforcement learning where the model leverages its internal adversarial feedback to iteratively improve anonymization performance. Experiments on two datasets show that our method outperforms baselines in both anonymization effectiveness (+12.3%) and utility (+6.8%). Our lightweight design supports direct deployment on edge devices, avoiding cloud reliance and communication-based privacy risks. Our code is open-source at https://github.com/tsinghua-fib-lab/AgentStealth. |
| 10 | DRARL: Disengagement-Reason-Augmented Reinforcement Learning for Efficient Improvement of Autonomous Driving Policy | 2025-06-20 | 展开With the increasing presence of automated vehicles on open roads under driver supervision, disengagement cases are becoming more prevalent. While some data-driven planning systems attempt to directly utilize these disengagement cases for policy improvement, the inherent scarcity of disengagement data (often occurring as a single instances) restricts training effectiveness. Furthermore, some disengagement data should be excluded since the disengagement may not always come from the failure of driving policies, e.g. the driver may casually intervene for a while. To this end, this work proposes disengagement-reason-augmented reinforcement learning (DRARL), which enhances driving policy improvement process according to the reason of disengagement cases. Specifically, the reason of disengagement is identified by a out-of-distribution (OOD) state estimation model. When the reason doesn't exist, the case will be identified as a casual disengagement case, which doesn't require additional policy adjustment. Otherwise, the policy can be updated under a reason-augmented imagination environment, improving the policy performance of disengagement cases with similar reasons. The method is evaluated using real-world disengagement cases collected by autonomous driving robotaxi. Experimental results demonstrate that the method accurately identifies policy-related disengagement reasons, allowing the agent to handle both original and semantically similar cases through reason-augmented training. Furthermore, the approach prevents the agent from becoming overly conservative after policy adjustments. Overall, this work provides an efficient way to improve driving policy performance with disengagement cases. |
| 11 | Investigating Memory in RL with POPGym Arcade | 2025-03-03 | 展开How should we analyze memory in deep RL? We introduce mathematical tools for fairly analyzing policies under partial observability and revealing how agents use memory to make decisions. To utilize these tools, we present POPGym Arcade, a collection of Atari-inspired, hardware-accelerated, pixel-based environments sharing a single observation and action space. Each environment provides fully and partially observable variants, enabling counterfactual studies on observability. We find that controlled studies are necessary for fair comparisons, and identify a pathology where value functions smear credit over irrelevant history. With this pathology, we demonstrate how out-of-distribution scenarios can contaminate memory, perturbing the policy far into the future, with implications for sim-to-real transfer and offline RL. |
| 12 | Multimodal Dreaming: A Global Workspace Approach to World Model-Based Reinforcement Learning | 2025-02-28 | 展开Humans leverage rich internal models of the world to reason about the future, imagine counterfactuals, and adapt flexibly to new situations. In Reinforcement Learning (RL), world models aim to capture how the environment evolves in response to the agent's actions, facilitating planning and generalization. However, typical world models directly operate on the environment variables (e.g. pixels, physical attributes), which can make their training slow and cumbersome; instead, it may be advantageous to rely on high-level latent dimensions that capture relevant multimodal variables. Global Workspace (GW) Theory offers a cognitive framework for multimodal integration and information broadcasting in the brain, and recent studies have begun to introduce efficient deep learning implementations of GW. Here, we evaluate the capabilities of an RL system combining GW with a world model. We compare our GW-Dreamer with various versions of the standard PPO and the original Dreamer algorithms. We show that performing the dreaming process (i.e., mental simulation) inside the GW latent space allows for training with fewer environment steps. As an additional emergent property, the resulting model (but not its comparison baselines) displays strong robustness to the absence of one of its observation modalities (images or simulation attributes). We conclude that the combination of GW with World Models holds great potential for improving decision-making in RL agents. |
| 13 | Online Decision MetaMorphFormer: A Casual Transformer-Based Reinforcement Learning Framework of Universal Embodied Intelligence | 2024-09-11 | 展开Interactive artificial intelligence in the motion control field is an interesting topic, especially when universal knowledge is adaptive to multiple tasks and universal environments. Despite there being increasing efforts in the field of Reinforcement Learning (RL) with the aid of transformers, most of them might be limited by the offline training pipeline, which prohibits exploration and generalization abilities. To address this limitation, we propose the framework of Online Decision MetaMorphFormer (ODM) which aims to achieve self-awareness, environment recognition, and action planning through a unified model architecture. Motivated by cognitive and behavioral psychology, an ODM agent is able to learn from others, recognize the world, and practice itself based on its own experience. ODM can also be applied to any arbitrary agent with a multi-joint body, located in different environments, and trained with different types of tasks using large-scale pre-trained datasets. Through the use of pre-trained datasets, ODM can quickly warm up and learn the necessary knowledge to perform the desired task, while the target environment continues to reinforce the universal policy. Extensive online experiments as well as few-shot and zero-shot environmental tests are used to verify ODM's performance and generalization ability. The results of our study contribute to the study of general artificial intelligence in embodied and cognitive fields. Code, results, and video examples can be found on the website \url{https://rlodm.github.io/odm/}. |
| 14 | Real-time Network Intrusion Detection via Decision Transformers | 2023-12-12 | 展开Many cybersecurity problems that require real-time decision-making based on temporal observations can be abstracted as a sequence modeling problem, e.g., network intrusion detection from a sequence of arriving packets. Existing approaches like reinforcement learning may not be suitable for such cybersecurity decision problems, since the Markovian property may not necessarily hold and the underlying network states are often not observable. In this paper, we cast the problem of real-time network intrusion detection as casual sequence modeling and draw upon the power of the transformer architecture for real-time decision-making. By conditioning a causal decision transformer on past trajectories, consisting of the rewards, network packets, and detection decisions, our proposed framework will generate future detection decisions to achieve the desired return. It enables decision transformers to be applied to real-time network intrusion detection, as well as a novel tradeoff between the accuracy and timeliness of detection. The proposed solution is evaluated on public network intrusion detection datasets and outperforms several baseline algorithms using reinforcement learning and sequence modeling, in terms of detection accuracy and timeliness. |
| 15 | Fine-Grained Session Recommendations in E-commerce using Deep Reinforcement Learning | 2022-10-20 | 展开Sustaining users' interest and keeping them engaged in the platform is very important for the success of an e-commerce business. A session encompasses different activities of a user between logging into the platform and logging out or making a purchase. User activities in a session can be classified into two groups: Known Intent and Unknown intent. Known intent activity pertains to the session where the intent of a user to browse/purchase a specific product can be easily captured. Whereas in unknown intent activity, the intent of the user is not known. For example, consider the scenario where a user enters the session to casually browse the products over the platform, similar to the window shopping experience in the offline setting. While recommending similar products is essential in the former, accurately understanding the intent and recommending interesting products is essential in the latter setting in order to retain a user. In this work, we focus primarily on the unknown intent setting where our objective is to recommend a sequence of products to a user in a session to sustain their interest, keep them engaged and possibly drive them towards purchase. We formulate this problem in the framework of the Markov Decision Process (MDP), a popular mathematical framework for sequential decision making and solve it using Deep Reinforcement Learning (DRL) techniques. However, training the next product recommendation is difficult in the RL paradigm due to large variance in browse/purchase behavior of the users. Therefore, we break the problem down into predicting various product attributes, where a pattern/trend can be identified and exploited to build accurate models. We show that the DRL agent provides better performance compared to a greedy strategy. |
| 16 | Adversarial Robustness of Deep Reinforcement Learning based Dynamic Recommender Systems | 2021-12-02 | 展开Adversarial attacks, e.g., adversarial perturbations of the input and adversarial samples, pose significant challenges to machine learning and deep learning techniques, including interactive recommendation systems. The latent embedding space of those techniques makes adversarial attacks difficult to detect at an early stage. Recent advance in causality shows that counterfactual can also be considered one of ways to generate the adversarial samples drawn from different distribution as the training samples. We propose to explore adversarial examples and attack agnostic detection on reinforcement learning-based interactive recommendation systems. We first craft different types of adversarial examples by adding perturbations to the input and intervening on the casual factors. Then, we augment recommendation systems by detecting potential attacks with a deep learning-based classifier based on the crafted data. Finally, we study the attack strength and frequency of adversarial examples and evaluate our model on standard datasets with multiple crafting methods. Our extensive experiments show that most adversarial attacks are effective, and both attack strength and attack frequency impact the attack performance. The strategically-timed attack achieves comparative attack performance with only 1/3 to 1/2 attack frequency. Besides, our black-box detector trained with one crafting method has the generalization ability over several other crafting methods. |
| 17 | Strategies for Using Proximal Policy Optimization in Mobile Puzzle Games | 2020-07-03 | 展开While traditionally a labour intensive task, the testing of game content is progressively becoming more automated. Among the many directions in which this automation is taking shape, automatic play-testing is one of the most promising thanks also to advancements of many supervised and reinforcement learning (RL) algorithms. However these type of algorithms, while extremely powerful, often suffer in production environments due to issues with reliability and transparency in their training and usage. In this research work we are investigating and evaluating strategies to apply the popular RL method Proximal Policy Optimization (PPO) in a casual mobile puzzle game with a specific focus on improving its reliability in training and generalization during game playing. We have implemented and tested a number of different strategies against a real-world mobile puzzle game (Lily's Garden from Tactile Games). We isolated the conditions that lead to a failure in either training or generalization during testing and we identified a few strategies to ensure a more stable behaviour of the algorithm in this game genre. |
| 18 | Causal Discovery with Reinforcement Learning | 2019-06-11 | 展开Discovering causal structure among a set of variables is a fundamental problem in many empirical sciences. Traditional score-based casual discovery methods rely on various local heuristics to search for a Directed Acyclic Graph (DAG) according to a predefined score function. While these methods, e.g., greedy equivalence search, may have attractive results with infinite samples and certain model assumptions, they are usually less satisfactory in practice due to finite data and possible violation of assumptions. Motivated by recent advances in neural combinatorial optimization, we propose to use Reinforcement Learning (RL) to search for the DAG with the best scoring. Our encoder-decoder model takes observable data as input and generates graph adjacency matrices that are used to compute rewards. The reward incorporates both the predefined score function and two penalty terms for enforcing acyclicity. In contrast with typical RL applications where the goal is to learn a policy, we use RL as a search strategy and our final output would be the graph, among all graphs generated during training, that achieves the best reward. We conduct experiments on both synthetic and real datasets, and show that the proposed approach not only has an improved search ability but also allows a flexible score function under the acyclicity constraint. |
| 序号 | 标题 | 日期 | 摘要 |
|---|---|---|---|
| 1 | A Design-Based Matching Framework for Staggered Adoption with Time-Varying Confounding | 2025-11-28 | 展开Causal inference in longitudinal datasets has long been challenging due to dynamic treatment adoption and confounding by time-varying covariates. Prior work either fails to account for heterogeneity across treatment adoption cohorts and treatment timings or relies on modeling assumptions. In this paper, we develop a novel design-based framework for inference on group- and time-specific treatment effects in panel data with staggered treatment adoption. We establish identification results for causal effects under this structure and introduce corresponding estimators, together with a block bootstrap procedure for estimating the covariance matrix and testing the homogeneity of group-time treatment effects. To implement the framework in practice, we propose the Reverse-Time Nested Matching algorithm, which constructs matched strata by pairing units from different adoption cohorts in a way that ensures comparability of covariate histories at each treatment time. Applying the algorithm to the Netflix-IPTV dataset, we find that while Netflix subscription does not significantly affect total IPTV viewing time, it does negatively affect VoD usage. We also provide statistical evidence that the causal effects of Netflix subscription may vary even within the same treatment cohort or across the same outcome and event times. |
| 2 | A General Bayesian Nonparametric Approach for Estimating Population-Level and Conditional Causal Effects | 2025-11-28 | 展开We propose a Bayesian nonparametric (BNP) approach to causal inference using observational data consisting of outcome, treatment, and a set of confounders. The conditional distribution of the outcome given treatment and confounders is modeled flexibly using a dependent nonparametric mixture model, in which both the atoms and the weights vary with the confounders. The proposed BNP model is well suited for causal inference problems, as it does not rely on parametric assumptions about how the conditional distribution depends on the confounders. In particular, the model effectively adjusts for confounding and improves the modeling of treatment effect heterogeneity, leading to more accurate estimation of both the average treatment effect (ATE) and heterogeneous treatment effects (HTE). Posterior inference under the proposed model is computationally efficient due to the use of data augmentation. Extensive evaluations demonstrate that the proposed model offers competitive or superior performance compared to a wide range of recent methods spanning various statistical approaches, including Bayesian additive regression tree (BART) models, which are well known for their strong empirical performance. More importantly, the model provides fully probabilistic inference on quantities of interest that other methods cannot easily provide, using their posterior distributions. |
| 3 | An empirical estimate of the electricity supply curve from market outcomes | 2025-11-28 | 展开Researchers and electricity sector practitioners frequently require the supply curve of electricity markets and the price elasticity of supply for purposes such as price forecasting, policy analyses or market power assessment. It is common practice to construct supply curves from engineering data such as installed capacity and fuel prices. In this study, we propose a data-driven methodology to estimate the supply curve of electricity market empirically, i.e. from observed prices and quantities without further modeling assumptions. Due to the massive swings in fuel prices during the European energy crisis, a central task is detecting periods of stable supply curves. To this end, we implement two alternative clustering methods, one based on the fundamental drivers of electricity supply and the other directly on observed market outcomes. We apply our methods to the German electricity market between 2019 and 2024. We find that both approaches identify almost identical regimes shifts, supporting the idea of stable supply regimes stemming from stable drivers. Supply conditions are often stable for extended periods, but evolved rapidly during the energy crisis, triggering a rapid succession of regimes. Fuel prices were the dominant drivers of regime shifts, while conventional plant availability and the nuclear phase-out play a comparatively minor role. Our approach produces empirical supply curves suitable for causal inference and counterfactual analysis of market outcomes. |
| 4 | Time Extrapolation with Graph Convolutional Autoencoder and Tensor Train Decomposition | 2025-11-28 | 展开Graph autoencoders have gained attention in nonlinear reduced-order modeling of parameterized partial differential equations defined on unstructured grids. Despite they provide a geometrically consistent way of treating complex domains, applying such architectures to parameterized dynamical systems for temporal prediction beyond the training data, i.e. the extrapolation regime, is still a challenging task due to the simultaneous need of temporal causality and generalizability in the parametric space. In this work, we explore the integration of graph convolutional autoencoders (GCAs) with tensor train (TT) decomposition and Operator Inference (OpInf) to develop a time-consistent reduced-order model. In particular, high-fidelity snapshots are represented as a combination of parametric, spatial, and temporal cores via TT decomposition, while OpInf is used to learn the evolution of the latter. Moreover, we enhance the generalization performance by developing a multi-fidelity two-stages approach in the framework of Deep Operator Networks (DeepONet), treating the spatial and temporal cores as the trunk networks, and the parametric core as the branch network. Numerical results, including heat-conduction, advection-diffusion and vortex-shedding phenomena, demonstrate great performance in effectively learning the dynamic in the extrapolation regime for complex geometries, also in comparison with state-of-the-art approaches e.g. MeshGraphNets. |
| 5 | Seeing before Observable: Potential Risk Reasoning in Autonomous Driving via Vision Language Models | 2025-11-28 | 展开Ensuring safety remains a key challenge for autonomous vehicles (AVs), especially in rare and complex scenarios. One critical but understudied aspect is the \textbf{potential risk} situations, where the risk is \textbf{not yet observable} but can be inferred from subtle precursors, such as anomalous behaviors or commonsense violations. Recognizing these precursors requires strong semantic understanding and reasoning capabilities, which are often absent in current AV systems due to the scarcity of such cases in existing driving or risk-centric datasets. Moreover, current autonomous driving accident datasets often lack annotations of the causal reasoning chains behind incidents, which are essential for identifying potential risks before they become observable. To address these gaps, we introduce PotentialRiskQA, a novel vision-language dataset designed for reasoning about potential risks prior to observation. Each sample is annotated with structured scene descriptions, semantic precursors, and inferred risk outcomes. Based on this dataset, we further propose PR-Reasoner, a vision-language-model-based framework tailored for onboard potential risk reasoning. Experimental results show that fine-tuning on PotentialRiskQA enables PR-Reasoner to significantly enhance its performance on the potential risk reasoning task compared to baseline VLMs. Together, our dataset and model provide a foundation for developing autonomous systems with improved foresight and proactive safety capabilities, moving toward more intelligent and resilient AVs. |
| 6 | CausalProfiler: Generating Synthetic Benchmarks for Rigorous and Transparent Evaluation of Causal Machine Learning | 2025-11-28 | 展开Causal machine learning (Causal ML) aims to answer "what if" questions using machine learning algorithms, making it a promising tool for high-stakes decision-making. Yet, empirical evaluation practices in Causal ML remain limited. Existing benchmarks often rely on a handful of hand-crafted or semi-synthetic datasets, leading to brittle, non-generalizable conclusions. To bridge this gap, we introduce CausalProfiler, a synthetic benchmark generator for Causal ML methods. Based on a set of explicit design choices about the class of causal models, queries, and data considered, the CausalProfiler randomly samples causal models, data, queries, and ground truths constituting the synthetic causal benchmarks. In this way, Causal ML methods can be rigorously and transparently evaluated under a variety of conditions. This work offers the first random generator of synthetic causal benchmarks with coverage guarantees and transparent assumptions operating on the three levels of causal reasoning: observation, intervention, and counterfactual. We demonstrate its utility by evaluating several state-of-the-art methods under diverse conditions and assumptions, both in and out of the identification regime, illustrating the types of analyses and insights the CausalProfiler enables. |
| 7 | Dark Speculation: Combining Qualitative and Quantitative Understanding in Frontier AI Risk Analysis | 2025-11-26 | 展开Estimating catastrophic harms from frontier AI is hindered by deep ambiguity: many of its risks are not only unobserved but unanticipated by analysts. The central limitation of current risk analysis is the inability to populate the |
| 8 | Context-Specific Causal Graph Discovery with Unobserved Contexts: Non-Stationarity, Regimes and Spatio-Temporal Patterns | 2025-11-26 | 展开Real-world data, for example in climate applications, often consists of spatially gridded time series data or data with comparable structure. While the underlying system is often believed to behave similar at different points in space and time, those variations that do exist are twofold relevant: They often encode important information in and of themselves. And they may negatively affect the stability / convergence and reliability\Slash{}validity of results of algorithms assuming stationarity or space-translation invariance. We study the information encoded in changes of the causal graph, with stability in mind. An analysis of this general task identifies two core challenges. We develop guiding principles to overcome these challenges, and provide a framework realizing these principles by modifying constraint-based causal discovery approaches on the level of independence testing. This leads to an extremely modular, easily extensible and widely applicable framework. It can leverage existing constraint-based causal discovery methods (demonstrated on IID-algorithms PC, PC-stable, FCI and time series algorithms PCMCI, PCMCI+, LPCMCI) with little to no modification. The built-in modularity allows to systematically understand and improve upon an entire array of subproblems. By design, it can be extended by leveraging insights from change-point-detection, clustering, independence-testing and other well-studied related problems. The division into more accessible sub-problems also simplifies the understanding of fundamental limitations, hyperparameters controlling trade-offs and the statistical interpretation of results. An open-source implementation will be available soon. |
| 9 | Physics Steering: Causal Control of Cross-Domain Concepts in a Physics Foundation Model | 2025-11-25 | 展开Recent advances in mechanistic interpretability have revealed that large language models (LLMs) develop internal representations corresponding not only to concrete entities but also distinct, human-understandable abstract concepts and behaviour. Moreover, these hidden features can be directly manipulated to steer model behaviour. However, it remains an open question whether this phenomenon is unique to models trained on inherently structured data (ie. language, images) or if it is a general property of foundation models. In this work, we investigate the internal representations of a large physics-focused foundation model. Inspired by recent work identifying single directions in activation space for complex behaviours in LLMs, we extract activation vectors from the model during forward passes over simulation datasets for different physical regimes. We then compute "delta" representations between the two regimes. These delta tensors act as concept directions in activation space, encoding specific physical features. By injecting these concept directions back into the model during inference, we can steer its predictions, demonstrating causal control over physical behaviours, such as inducing or removing some particular physical feature from a simulation. These results suggest that scientific foundation models learn generalised representations of physical principles. They do not merely rely on superficial correlations and patterns in the simulations. Our findings open new avenues for understanding and controlling scientific foundation models and has implications for AI-enabled scientific discovery. |
| 10 | Spatio-Temporal Hierarchical Causal Models | 2025-11-25 | 展开The abundance of fine-grained spatio-temporal data, such as traffic sensor networks, offers vast opportunities for scientific discovery. However, inferring causal relationships from such observational data remains challenging, particularly due to unobserved confounders that are specific to units (e.g., geographical locations) yet influence outcomes over time. Most existing methods for spatio-temporal causal inference assume that all confounders are observed, an assumption that is often violated in practice. In this paper, we introduce Spatio-Temporal Hierarchical Causal Models (ST-HCMs), a novel graphical framework that extends hierarchical causal modeling to the spatio-temporal domain. At the core of our approach is the Spatio-Temporal Collapse Theorem, which shows that a complex ST-HCM converges to a simpler flat causal model as the amount of subunit data increases. This theoretical result enables a general procedure for causal identification, allowing ST-HCMs to recover causal effects even in the presence of unobserved, time-invariant unit-level confounders, a scenario where standard non-hierarchical models fail. We validate the effectiveness of our framework on both synthetic and real-world datasets, demonstrating its potential for robust causal inference in complex dynamic systems. |
| 11 | Learning Subgroups with Maximum Treatment Effects without Causal Heuristics | 2025-11-25 | 展开Discovering subgroups with the maximum average treatment effect is crucial for targeted decision making in domains such as precision medicine, public policy, and education. While most prior work is formulated in the potential outcome framework, the corresponding structural causal model (SCM) for this task has been largely overlooked. In practice, two approaches dominate. The first estimates pointwise conditional treatment effects and then fits a tree on those estimates, effectively turning subgroup estimation into the harder problem of accurate pointwise estimation. The second constructs decision trees or rule sets with ad-hoc 'causal' heuristics, typically without rigorous justification for why a given heuristic may be used or whether such heuristics are necessary at all. We address these issues by studying the problem directly under the SCM framework. Under the assumption of a partition-based model, we show that optimal subgroup discovery reduces to recovering the data-generating models and hence a standard supervised learning problem (regression or classification). This allows us to adopt any partition-based methods to learn the subgroup from data. We instantiate the approach with CART, arguably one of the most widely used tree-based methods, to learn the subgroup with maximum treatment effect. Finally, on a large collection of synthetic and semi-synthetic datasets, we compare our method against a wide range of baselines and find that our approach, which avoids such causal heuristics, more accurately identifies subgroups with maximum treatment effect. Our source code is available at https://github.com/ylincen/causal-subgroup. |
| 12 | CounterVQA: Evaluating and Improving Counterfactual Reasoning in Vision-Language Models for Video Understanding | 2025-11-25 | 展开Vision Language Models (VLMs) have recently shown significant advancements in video understanding, especially in feature alignment, event reasoning, and instruction-following tasks. However, their capability for counterfactual reasoning, inferring alternative outcomes under hypothetical conditions, remains underexplored. This capability is essential for robust video understanding, as it requires identifying underlying causal structures and reasoning about unobserved possibilities, rather than merely recognizing observed patterns. To systematically evaluate this capability, we introduce CounterVQA, a video-based benchmark featuring three progressive difficulty levels that assess different aspects of counterfactual reasoning. Through comprehensive evaluation of both state-of-the-art open-source and closed-source models, we uncover a substantial performance gap: while these models achieve reasonable accuracy on simple counterfactual questions, performance degrades significantly on complex multi-hop causal chains. To address these limitations, we develop a post-training method, CFGPT, that enhances a model's visual counterfactual reasoning ability by distilling its counterfactual reasoning capability from the language modality, yielding consistent improvements across all CounterVQA difficulty levels. Dataset and code will be further released. |
| 13 | Scalable Bayesian Network Structure Learning Using Tsetlin Machine to Constrain the Search Space | 2025-11-24 | 展开The PC algorithm is a widely used method in causal inference for learning the structure of Bayesian networks. Despite its popularity, the PC algorithm suffers from significant time complexity, particularly as the size of the dataset increases, which limits its applicability in large-scale real-world problems. In this study, we propose a novel approach that utilises the Tsetlin Machine (TM) to construct Bayesian structures more efficiently. Our method leverages the most significant literals extracted from the TM and performs conditional independence (CI) tests on these selected literals instead of the full set of variables, resulting in a considerable reduction in computational time. We implemented our approach and compared it with various state-of-the-art methods. Our evaluation includes categorical datasets from the bnlearn repository, such as Munin1, Hepar2. The findings indicate that the proposed TM-based method not only reduces computational complexity but also maintains competitive accuracy in causal discovery, making it a viable alternative to traditional PC algorithm implementations by offering improved efficiency without compromising performance. |
| 14 | Fidelity-Aware Recommendation Explanations via Stochastic Path Integration | 2025-11-22 | 展开Explanation fidelity, which measures how accurately an explanation reflects a model's true reasoning, remains critically underexplored in recommender systems. We introduce SPINRec (Stochastic Path Integration for Neural Recommender Explanations), a model-agnostic approach that adapts path-integration techniques to the sparse and implicit nature of recommendation data. To overcome the limitations of prior methods, SPINRec employs stochastic baseline sampling: instead of integrating from a fixed or unrealistic baseline, it samples multiple plausible user profiles from the empirical data distribution and selects the most faithful attribution path. This design captures the influence of both observed and unobserved interactions, yielding more stable and personalized explanations. We conduct the most comprehensive fidelity evaluation to date across three models (MF, VAE, NCF), three datasets (ML1M, Yahoo! Music, Pinterest), and a suite of counterfactual metrics, including AUC-based perturbation curves and fixed-length diagnostics. SPINRec consistently outperforms all baselines, establishing a new benchmark for faithful explainability in recommendation. Code and evaluation tools are publicly available at https://github.com/DeltaLabTLV/SPINRec. |
| 15 | Counterfactual World Models via Digital Twin-conditioned Video Diffusion | 2025-11-21 | 展开World models learn to predict the temporal evolution of visual observations given a control signal, potentially enabling agents to reason about environments through forward simulation. Because of the focus on forward simulation, current world models generate predictions based on factual observations. For many emerging applications, such as comprehensive evaluations of physical AI behavior under varying conditions, the ability of world models to answer counterfactual queries, such as "what would happen if this object was removed?", is of increasing importance. We formalize counterfactual world models that additionally take interventions as explicit inputs, predicting temporal sequences under hypothetical modifications to observed scene properties. Traditional world models operate directly on entangled pixel-space representations where object properties and relationships cannot be selectively modified. This modeling choice prevents targeted interventions on specific scene properties. We introduce CWMDT, a framework to overcome those limitations, turning standard video diffusion models into effective counterfactual world models. First, CWMDT constructs digital twins of observed scenes to explicitly encode objects and their relationships, represented as structured text. Second, CWMDT applies large language models to reason over these representations and predict how a counterfactual intervention propagates through time to alter the observed scene. Third, CWMDT conditions a video diffusion model with the modified representation to generate counterfactual visual sequences. Evaluations on two benchmarks show that the CWMDT approach achieves state-of-the-art performance, suggesting that alternative representations of videos, such as the digital twins considered here, offer powerful control signals for video forward simulation-based world models. |
| 16 | Beyond Generative AI: World Models for Clinical Prediction, Counterfactuals, and Planning | 2025-11-20 | 展开Healthcare requires AI that is predictive, reliable, and data-efficient. However, recent generative models lack physical foundation and temporal reasoning required for clinical decision support. As scaling language models show diminishing returns for grounded clinical reasoning, world models are gaining traction because they learn multimodal, temporally coherent, and action-conditioned representations that reflect the physical and causal structure of care. This paper reviews World Models for healthcare systems that learn predictive dynamics to enable multistep rollouts, counterfactual evaluation and planning. We survey recent work across three domains: (i) medical imaging and diagnostics (e.g., longitudinal tumor simulation, projection-transition modeling, and Joint Embedding Predictive Architecture i.e., JEPA-style predictive representation learning), (ii) disease progression modeling from electronic health records (generative event forecasting at scale), and (iii) robotic surgery and surgical planning (action-conditioned guidance and control). We also introduce a capability rubric: L1 temporal prediction, L2 action-conditioned prediction, L3 counterfactual rollouts for decision support, and L4 planning/control. Most reviewed systems achieve L1--L2, with fewer instances of L3 and rare L4. We identify cross-cutting gaps that limit clinical reliability; under-specified action spaces and safety constraints, weak interventional validation, incomplete multimodal state construction, and limited trajectory-level uncertainty calibration. This review outlines a research agenda for clinically robust prediction-first world models that integrate generative backbones (transformers, diffusion, VAE) with causal/mechanical foundation for safe decision support in healthcare. |
| 17 | Error-Driven Scene Editing for 3D Grounding in Large Language Models | 2025-11-18 | 展开Despite recent progress in 3D-LLMs, they remain limited in accurately grounding language to visual and spatial elements in 3D environments. This limitation stems in part from training data that focuses on language reasoning rather than spatial understanding due to scarce 3D resources, leaving inherent grounding biases unresolved. To address this, we propose 3D scene editing as a key mechanism to generate precise visual counterfactuals that mitigate these biases through fine-grained spatial manipulation, without requiring costly scene reconstruction or large-scale 3D data collection. Furthermore, to make these edits targeted and directly address the specific weaknesses of the model, we introduce DEER-3D, an error-driven framework following a structured "Decompose, Diagnostic Evaluation, Edit, and Re-train" workflow, rather than broadly or randomly augmenting data as in conventional approaches. Specifically, upon identifying a grounding failure of the 3D-LLM, our framework first diagnoses the exact predicate-level error (e.g., attribute or spatial relation). It then executes minimal, predicate-aligned 3D scene edits, such as recoloring or repositioning, to produce targeted counterfactual supervision for iterative model fine-tuning, significantly enhancing grounding accuracy. We evaluate our editing pipeline across multiple benchmarks for 3D grounding and scene understanding tasks, consistently demonstrating improvements across all evaluated datasets through iterative refinement. DEER-3D underscores the effectiveness of targeted, error-driven scene editing in bridging linguistic reasoning capabilities with spatial grounding in 3D LLMs. |
| 18 | Comment on "Repair of DNA Double-Strand Breaks Leaves Heritable Impairment to Genome Function" | 2025-11-15 | 展开Bantele and colleagues recently reported that repair of a single CRISPR/Cas9-induced DNA double-strand break (DSB) in the c-MYC topologically associated domain leads to a persistent depletion of chromatin interactions and long-term transcriptional attenuation across multiple generations of human cells. They interpret this observation as evidence for a previously unrecognized principle--"chromatin fatigue"--in which DSB repair generates a stable architectural defect that acts as a heritable impairment to genome function. Such an idea, if correct, would carry profound implications for genome biology, epigenetic inheritance, cancer evolution, aging, and the safety of therapeutic genome editing. However, our detailed reassessment of the experimental design, underlying assumptions, and data interpretation reveals that the evidence provided is inadequate to support these sweeping conclusions. Instead, the observed outcomes are more plausibly explained by a combination of Cas9 persistence, off-target DNA damage, repair-factor retention, MYC enhancer plasticity, and the well-documented genomic instability of HeLa cells. The study does not demonstrate mechanistic causality, does not exclude simpler explanations, and does not provide data consistent with true chromatin memory or heritable architectural change. Moreover, its statistical inferences are based on noisy measurements that fall within expected variability of unstable oncogenic loci. Here, we present a comprehensive critical analysis showing that the proposed model of chromatin fatigue is unsupported by the available evidence. We offer a corrected interpretation in which the chromatin landscape experiences a temporary, repair-associated perturbation that resolves without leaving enduring or heritable impairment. |
| 19 | A Conditional Distribution Equality Testing Framework using Deep Generative Learning | 2025-09-22 | 展开In this paper, we propose a general framework for testing the conditional distribution equality in a two-sample problem, which is most relevant to covariate shift and causal discovery. Our framework is built on neural network-based generative methods and sample splitting techniques by transforming the conditional testing problem into an unconditional one. We introduce the generative classification accuracy-based conditional distribution equality test (GCA-CDET) to illustrate the proposed framework. We establish the convergence rate for the learned generator by deriving new results related to the recently-developed offset Rademacher complexity and prove the testing consistency of GCA-CDET under mild conditions.Empirically, we conduct numerical studies including synthetic datasets and two real-world datasets, demonstrating the effectiveness of our approach. Additional discussions on the optimality of the proposed framework are provided in the online supplementary material. |
| 20 | CAMA: Enhancing Mathematical Reasoning in Large Language Models with Causal Knowledge | 2025-08-04 | 展开Large Language Models (LLMs) have demonstrated strong performance across a wide range of tasks, yet they still struggle with complex mathematical reasoning, a challenge fundamentally rooted in deep structural dependencies. To address this challenge, we propose \textbf{CA}usal \textbf{MA}thematician (\textbf{CAMA}), a two-stage causal framework that equips LLMs with explicit, reusable mathematical structure. In the learning stage, CAMA first constructs the \textbf{M}athematical \textbf{C}ausal \textbf{G}raph (\textbf{MCG}), a high-level representation of solution strategies, by combining LLM priors with causal discovery algorithms applied to a corpus of question-solution pairs. The resulting MCG encodes essential knowledge points and their causal dependencies. To better align the graph with downstream reasoning tasks, CAMA further refines the MCG through iterative feedback derived from a selected subset of the question-solution pairs. In the reasoning stage, given a new question, CAMA dynamically extracts a task-relevant subgraph from the MCG, conditioned on both the question content and the LLM's intermediate reasoning trace. This subgraph, which encodes the most pertinent knowledge points and their causal dependencies, is then injected back into the LLM to guide its reasoning process. Empirical results on real-world datasets show that CAMA significantly improves LLM performance on challenging mathematical problems. Furthermore, our experiments demonstrate that structured guidance consistently outperforms unstructured alternatives, and that incorporating asymmetric causal relationships yields greater improvements than using symmetric associations alone. |
| 21 | Estimating Interpretable Heterogeneous Treatment Effect with Causal Subgroup Discovery in Survival Outcomes | 2024-09-28 | 展开Estimating heterogeneous treatment effect (HTE) for survival outcomes has gained increasing attention, as it captures the variation in treatment efficacy across patients or subgroups in delaying disease progression. However, most existing methods focus on post-hoc subgroup identification rather than simultaneously estimating HTE and selecting relevant subgroups. In this paper, we propose an interpretable HTE estimation framework that integrates three meta-learners that simultaneously estimate CATE for survival outcomes and identify predictive subgroups. We evaluated the performance of our method through comprehensive simulation studies across various randomized clinical trial (RCT) settings. Additionally, we demonstrated its application in a large RCT for age-related macular degeneration (AMD), a polygenic progressive eye disease, to estimate the HTE of an antioxidant and mineral supplement on time-to-AMD progression and to identify genetics-based subgroups with enhanced treatment effects. Our method offers a direct interpretation of the estimated HTE and provides evidence to support precision healthcare. |
| 序号 | 标题 | 日期 | 摘要 |
|---|---|---|---|
| 1 | HunyuanVideo 1.5 Technical Report | 2025-11-24 | 展开We present HunyuanVideo 1.5, a lightweight yet powerful open-source video generation model that achieves state-of-the-art visual quality and motion coherence with only 8.3 billion parameters, enabling efficient inference on consumer-grade GPUs. This achievement is built upon several key components, including meticulous data curation, an advanced DiT architecture featuring selective and sliding tile attention (SSTA), enhanced bilingual understanding through glyph-aware text encoding, progressive pre-training and post-training, and an efficient video super-resolution network. Leveraging these designs, we developed a unified framework capable of high-quality text-to-video and image-to-video generation across multiple durations and resolutions. Extensive experiments demonstrate that this compact and proficient model establishes a new state-of-the-art among open-source video generation models. By releasing the code and model weights, we provide the community with a high-performance foundation that lowers the barrier to video creation and research, making advanced video generation accessible to a broader audience. All open-source assets are publicly available at https://github.com/Tencent-Hunyuan/HunyuanVideo-1.5. |
| 2 | STCDiT: Spatio-Temporally Consistent Diffusion Transformer for High-Quality Video Super-Resolution | 2025-11-24 | 展开We present STCDiT, a video super-resolution framework built upon a pre-trained video diffusion model, aiming to restore structurally faithful and temporally stable videos from degraded inputs, even under complex camera motions. The main challenges lie in maintaining temporal stability during reconstruction and preserving structural fidelity during generation. To address these challenges, we first develop a motion-aware VAE reconstruction method that performs segment-wise reconstruction, with each segment clip exhibiting uniform motion characteristic, thereby effectively handling videos with complex camera motions. Moreover, we observe that the first-frame latent extracted by the VAE encoder in each clip, termed the anchor-frame latent, remains unaffected by temporal compression and retains richer spatial structural information than subsequent frame latents. We further develop an anchor-frame guidance approach that leverages structural information from anchor frames to constrain the generation process and improve structural fidelity of video features. Coupling these two designs enables the video diffusion model to achieve high-quality video super-resolution. Extensive experiments show that STCDiT outperforms state-of-the-art methods in terms of structural fidelity and temporal consistency. |
| 3 | Rethinking Diffusion Model-Based Video Super-Resolution: Leveraging Dense Guidance from Aligned Features | 2025-11-21 | 展开Diffusion model (DM) based Video Super-Resolution (VSR) approaches achieve impressive perceptual quality. However, they suffer from error accumulation, spatial artifacts, and a trade-off between perceptual quality and fidelity, primarily caused by inaccurate alignment and insufficient compensation between video frames. In this paper, within the DM-based VSR pipeline, we revisit the role of alignment and compensation between adjacent video frames and reveal two crucial observations: (a) the feature domain is better suited than the pixel domain for information compensation due to its stronger spatial and temporal correlations, and (b) warping at an upscaled resolution better preserves high-frequency information, but this benefit is not necessarily monotonic. Therefore, we propose a novel Densely Guided diffusion model with Aligned Features for Video Super-Resolution (DGAF-VSR), with an Optical Guided Warping Module (OGWM) to maintain high-frequency details in the aligned features and a Feature-wise Temporal Condition Module (FTCM) to deliver dense guidance in the feature domain. Extensive experiments on synthetic and real-world datasets demonstrate that DGAF-VSR surpasses state-of-the-art methods in key aspects of VSR, including perceptual quality (35.82% DISTS reduction), fidelity (0.20 dB PSNR gain), and temporal consistency (30.37% tLPIPS reduction). |
| 4 | InstantViR: Real-Time Video Inverse Problem Solver with Distilled Diffusion Prior | 2025-11-18 | 展开Video inverse problems are fundamental to streaming, telepresence, and AR/VR, where high perceptual quality must coexist with tight latency constraints. Diffusion-based priors currently deliver state-of-the-art reconstructions, but existing approaches either adapt image diffusion models with ad hoc temporal regularizers - leading to temporal artifacts - or rely on native video diffusion models whose iterative posterior sampling is far too slow for real-time use. We introduce InstantViR, an amortized inference framework for ultra-fast video reconstruction powered by a pre-trained video diffusion prior. We distill a powerful bidirectional video diffusion model (teacher) into a causal autoregressive student that maps a degraded video directly to its restored version in a single forward pass, inheriting the teacher's strong temporal modeling while completely removing iterative test-time optimization. The distillation is prior-driven: it only requires the teacher diffusion model and known degradation operators, and does not rely on externally paired clean/noisy video data. To further boost throughput, we replace the video-diffusion backbone VAE with a high-efficiency LeanVAE via an innovative teacher-space regularized distillation scheme, enabling low-latency latent-space processing. Across streaming random inpainting, Gaussian deblurring and super-resolution, InstantViR matches or surpasses the reconstruction quality of diffusion-based baselines while running at over 35 FPS on NVIDIA A100 GPUs, achieving up to 100 times speedups over iterative video diffusion solvers. These results show that diffusion-based video reconstruction is compatible with real-time, interactive, editable, streaming scenarios, turning high-quality video restoration into a practical component of modern vision systems. |
| 5 | VividFace: High-Quality and Efficient One-Step Diffusion For Video Face Enhancement | 2025-09-28 | 展开Video Face Enhancement (VFE) aims to restore high-quality facial regions from degraded video sequences, enabling a wide range of practical applications. Despite substantial progress in the field, current methods that primarily rely on video super-resolution and generative frameworks continue to face three fundamental challenges: (1) computational inefficiency caused by iterative multi-step denoising in diffusion models; (2) faithfully modeling intricate facial textures while preserving temporal consistency; and (3) limited model generalization due to the lack of high-quality face video training data. To address these challenges, we propose VividFace, a novel and efficient one-step diffusion framework for VFE. Built upon the pretrained WANX video generation model, VividFace reformulates the traditional multi-step diffusion process as a single-step flow matching paradigm that directly maps degraded inputs to high-quality outputs with significantly reduced inference time. To enhance facial detail recovery, we introduce a Joint Latent-Pixel Face-Focused Training strategy that constructs spatiotemporally aligned facial masks to guide optimization toward critical facial regions in both latent and pixel spaces. Furthermore, we develop an MLLM-driven automated filtering pipeline that produces MLLM-Face90, a meticulously curated high-quality face video dataset, ensuring models learn from photorealistic facial textures. Extensive experiments demonstrate that VividFace achieves superior performance in perceptual quality, identity preservation, and temporal consistency across both synthetic and real-world benchmarks. We will publicly release our code, models, and dataset to support future research. |
| 6 | SpatioTemporal Difference Network for Video Depth Super-Resolution | 2025-08-02 | 展开Depth super-resolution has achieved impressive performance, and the incorporation of multi-frame information further enhances reconstruction quality. Nevertheless, statistical analyses reveal that video depth super-resolution remains affected by pronounced long-tailed distributions, with the long-tailed effects primarily manifesting in spatial non-smooth regions and temporal variation zones. To address these challenges, we propose a novel SpatioTemporal Difference Network (STDNet) comprising two core branches: a spatial difference branch and a temporal difference branch. In the spatial difference branch, we introduce a spatial difference mechanism to mitigate the long-tailed issues in spatial non-smooth regions. This mechanism dynamically aligns RGB features with learned spatial difference representations, enabling intra-frame RGB-D aggregation for depth calibration. In the temporal difference branch, we further design a temporal difference strategy that preferentially propagates temporal variation information from adjacent RGB and depth frames to the current depth frame, leveraging temporal difference representations to achieve precise motion compensation in temporal long-tailed areas. Extensive experimental results across multiple datasets demonstrate the effectiveness of our STDNet, outperforming existing approaches. |
| 7 | BusterX: MLLM-Powered AI-Generated Video Forgery Detection and Explanation | 2025-05-19 | 展开Advances in AI generative models facilitate super-realistic video synthesis, amplifying misinformation risks via social media and eroding trust in digital content. Several research works have explored new deepfake detection methods on AI-generated images to alleviate these risks. However, with the fast development of video generation models, such as Sora and WanX, there is currently a lack of large-scale, high-quality AI-generated video datasets for forgery detection. In addition, existing detection approaches predominantly treat the task as binary classification, lacking explainability in model decision-making and failing to provide actionable insights or guidance for the public. To address these challenges, we propose \textbf{GenBuster-200K}, a large-scale AI-generated video dataset featuring 200K high-resolution video clips, diverse latest generative techniques, emphasis on fairness, and focus on real-world scenes. We further introduce \textbf{BusterX}, a novel AI-generated video detection and explanation framework leveraging multimodal large language model (MLLM) and reinforcement learning (RL) to provide authenticity determination and explainable rationales. To our knowledge, BusterX is the first framework to integrate MLLM with RL for explainable AI-generated video detection. Extensive experiments with state-of-the-art methods and ablation studies demonstrate the effectiveness and generalizability of BusterX. |
| 8 | Temporal Inconsistency Guidance for Super-resolution Video Quality Assessment | 2024-12-25 | 展开As super-resolution (SR) techniques introduce unique distortions that fundamentally differ from those caused by traditional degradation processes (e.g., compression), there is an increasing demand for specialized video quality assessment (VQA) methods tailored to SR-generated content. One critical factor affecting perceived quality is temporal inconsistency, which refers to irregularities between consecutive frames. However, existing VQA approaches rarely quantify this phenomenon or explicitly investigate its relationship with human perception. Moreover, SR videos exhibit amplified inconsistency levels as a result of enhancement processes. In this paper, we propose \textit{Temporal Inconsistency Guidance for Super-resolution Video Quality Assessment (TIG-SVQA)} that underscores the critical role of temporal inconsistency in guiding the quality assessment of SR videos. We first design a perception-oriented approach to quantify frame-wise temporal inconsistency. Based on this, we introduce the Inconsistency Highlighted Spatial Module, which localizes inconsistent regions at both coarse and fine scales. Inspired by the human visual system, we further develop an Inconsistency Guided Temporal Module that performs progressive temporal feature aggregation: (1) a consistency-aware fusion stage in which a visual memory capacity block adaptively determines the information load of each temporal segment based on inconsistency levels, and (2) an informative filtering stage for emphasizing quality-related features. Extensive experiments on both single-frame and multi-frame SR video scenarios demonstrate that our method significantly outperforms state-of-the-art VQA approaches. The code is publicly available at https://github.com/Lighting-YXLI/TIG-SVQA-main. |
7. knowledge graph/knowledge distillation/knowledge representation/knowledge transfer/knowledge embedding
| 序号 | 标题 | 日期 | 摘要 |
|---|---|---|---|
| 1 | LFM2 Technical Report | 2025-11-28 | 展开We present LFM2, a family of Liquid Foundation Models designed for efficient on-device deployment and strong task capabilities. Using hardware-in-the-loop architecture search under edge latency and memory constraints, we obtain a compact hybrid backbone that combines gated short convolutions with a small number of grouped query attention blocks, delivering up to 2x faster prefill and decode on CPUs compared to similarly sized models. The LFM2 family covers 350M-8.3B parameters, including dense models (350M, 700M, 1.2B, 2.6B) and a mixture-of-experts variant (8.3B total, 1.5B active), all with 32K context length. LFM2's training pipeline includes a tempered, decoupled Top-K knowledge distillation objective that avoids support mismatch; curriculum learning with difficulty-ordered data; and a three-stage post-training recipe of supervised fine-tuning, length-normalized preference optimization, and model merging. Pre-trained on 10-12T tokens, LFM2 models achieve strong results across diverse benchmarks; for example, LFM2-2.6B reaches 79.56% on IFEval and 82.41% on GSM8K. We further build multimodal and retrieval variants: LFM2-VL for vision-language tasks, LFM2-Audio for speech, and LFM2-ColBERT for retrieval. LFM2-VL supports tunable accuracy-latency tradeoffs via token-efficient visual processing, while LFM2-Audio separates audio input and output pathways to enable real-time speech-to-speech interaction competitive with models 3x larger. LFM2-ColBERT provides a low-latency encoder for queries and documents, enabling high-performance retrieval across multiple languages. All models are released with open weights and deployment packages for ExecuTorch, llama.cpp, and vLLM, making LFM2 a practical base for edge applications that need fast, memory-efficient inference and strong task capabilities. |
| 2 | Towards Improving Interpretability of Language Model Generation through a Structured Knowledge Discovery Approach | 2025-11-28 | 展开Knowledge-enhanced text generation aims to enhance the quality of generated text by utilizing internal or external knowledge sources. While language models have demonstrated impressive capabilities in generating coherent and fluent text, the lack of interpretability presents a substantial obstacle. The limited interpretability of generated text significantly impacts its practical usability, particularly in knowledge-enhanced text generation tasks that necessitate reliability and explainability. Existing methods often employ domain-specific knowledge retrievers that are tailored to specific data characteristics, limiting their generalizability to diverse data types and tasks. To overcome this limitation, we directly leverage the two-tier architecture of structured knowledge, consisting of high-level entities and low-level knowledge triples, to design our task-agnostic structured knowledge hunter. Specifically, we employ a local-global interaction scheme for structured knowledge representation learning and a hierarchical transformer-based pointer network as the backbone for selecting relevant knowledge triples and entities. By combining the strong generative ability of language models with the high faithfulness of the knowledge hunter, our model achieves high interpretability, enabling users to comprehend the model output generation process. Furthermore, we empirically demonstrate the effectiveness of our model in both internal knowledge-enhanced table-to-text generation on the RotoWireFG dataset and external knowledge-enhanced dialogue response generation on the KdConv dataset. Our task-agnostic model outperforms state-of-the-art methods and corresponding language models, setting new standards on the benchmark. |
| 3 | Beyond 2-Edge-Connectivity: Algorithms and Impossibility for Content-Oblivious Leader Election | 2025-11-28 | 展开The content-oblivious model, introduced by Censor-Hillel, Cohen, Gelles, and Sel (PODC 2022; Distributed Computing 2023), captures an extremely weak form of communication where nodes can only send asynchronous, content-less pulses. Censor-Hillel, Cohen, Gelles, and Sel showed that no non-constant function |
| 4 | Improved and Parameterized Algorithms for Online Multi-level Aggregation: A Memory-based Approach | 2025-11-28 | 展开We study the online multi-level aggregation problem with deadlines (MLAP-D) introduced by Bienkowski et al. (ESA 2016, OR 2020). In this problem, requests arrive over time at the vertices of a given vertex-weighted tree, and each request has a deadline that it must be served by. The cost of serving a request equals the cost of a path from the root to the vertex where the request resides. Instead of serving each request individually, requests can be aggregated and served by transmitting a subtree from the root that spans the vertices on which the requests reside, to potentially be more cost-effective. The aggregated cost is the weight of the transmission subtree. The goal of MLAP-D is to find an aggregation solution that minimizes the total cost while serving all requests. We present improved and parameterized algorithms for MLAP-D. Our result is twofold. First, we present an |
| 5 | Pathryoshka: Compressing Pathology Foundation Models via Multi-Teacher Knowledge Distillation with Nested Embeddings | 2025-11-28 | 展开Pathology foundation models (FMs) have driven significant progress in computational pathology. However, these high-performing models can easily exceed a billion parameters and produce high-dimensional embeddings, thus limiting their applicability for research or clinical use when computing resources are tight. Here, we introduce Pathryoshka, a multi-teacher distillation framework inspired by RADIO distillation and Matryoshka Representation Learning to reduce pathology FM sizes while allowing for adaptable embedding dimensions. We evaluate our framework with a distilled model on ten public pathology benchmarks with varying downstream tasks. Compared to its much larger teachers, Pathryoshka reduces the model size by 86-92% at on-par performance. It outperforms state-of-the-art single-teacher distillation models of comparable size by a median margin of 7.0 in accuracy. By enabling efficient local deployment without sacrificing accuracy or representational richness, Pathryoshka democratizes access to state-of-the-art pathology FMs for the broader research and clinical community. |
| 6 | Identification of Malicious Posts on the Dark Web Using Supervised Machine Learning | 2025-11-28 | 展开Given the constant growth and increasing sophistication of cyberattacks, cybersecurity can no longer rely solely on traditional defense techniques and tools. Proactive detection of cyber threats has become essential to help security teams identify potential risks and implement effective mitigation measures. Cyber Threat Intelligence (CTI) plays a key role by providing security analysts with evidence-based knowledge about cyber threats. CTI information can be extracted using various techniques and data sources; however, machine learning has proven promising. As for data sources, social networks and online discussion forums are commonly explored. In this study, we apply text mining techniques and machine learning to data collected from Dark Web forums in Brazilian Portuguese to identify malicious posts. Our contributions include the creation of three original datasets, a novel multi-stage labeling process combining indicators of compromise (IoCs), contextual keywords, and manual analysis, and a comprehensive evaluation of text representations and classifiers. To our knowledge, this is the first study to focus specifically on Brazilian Portuguese content in this domain. The best-performing model, using LightGBM and TF-IDF, was able to detect relevant posts with high accuracy. We also applied topic modeling to validate the model's outputs on unlabeled data, confirming its robustness in real-world scenarios. |
| 7 | Analyzing Image Beyond Visual Aspect: Image Emotion Classification via Multiple-Affective Captioning | 2025-11-28 | 展开Image emotion classification (IEC) is a longstanding research field that has received increasing attention with the rapid progress of deep learning. Although recent advances have leveraged the knowledge encoded in pre-trained visual models, their effectiveness is constrained by the "affective gap" , limits the applicability of pre-training knowledge for IEC tasks. It has been demonstrated in psychology that language exhibits high variability, encompasses diverse and abundant information, and can effectively eliminate the "affective gap". Inspired by this, we propose a novel Affective Captioning for Image Emotion Classification (ACIEC) to classify image emotion based on pure texts, which effectively capture the affective information in the image. In our method, a hierarchical multi-level contrastive loss is designed for detecting emotional concepts from images, while an emotional attribute chain-of-thought reasoning is proposed to generate affective sentences. Then, a pre-trained language model is leveraged to synthesize emotional concepts and affective sentences to conduct IEC. Additionally, a contrastive loss based on semantic similarity sampling is designed to solve the problem of large intra-class differences and small inter-class differences in affective datasets. Moreover, we also take the images with embedded texts into consideration, which were ignored by previous studies. Extensive experiments illustrate that our method can effectively bridge the affective gap and achieve superior results on multiple benchmarks. |
| 8 | Identification of contractive Lur'e-type systems via kernel-based Lipschitz design | 2025-11-28 | 展开This paper addresses the problem of identifying contractive Lur'e-type systems. Specifically, it proposes an identification framework that integrates linear prior knowledge with a kernel representation of the nonlinear feedback while systematically enforcing contractivity via Lipschitz constant design. The resulting algorithms provide models that are accurate in prediction, interpretable, and faithful to the contractive nature of the true system. Numerical experiments demonstrate that enforcing contractivity significantly improves parameter estimation and yields models that are both accurate and physically meaningful. |
| 9 | Benchmarking neutral atom-based quantum processors at scale | 2025-11-28 | 展开In recent years, neutral atom-based quantum computation has been established as a competing alternative for the realization of fault-tolerant quantum computation. However, as with other quantum technologies, various sources of noise limit their performance. With processors continuing to scale up, new techniques are needed to characterize and compare them in order to track their progress. In this work, we present two systematic benchmarks that evaluate these quantum processors at scale. We use the quantum adiabatic algorithm (QAA) and the quantum approximate optimization algorithm (QAOA) to solve maximal independent set (MIS) instances of random unit-disk graphs. These benchmarks are scalable, relying not on prior knowledge of the system's evolution but on the quality of the MIS solutions obtained. We benchmark quera_aquila and pasqal_fresnel on problem sizes up to 102 and 85 qubits, respectively. Overall, quera_aquila performs better on QAOA and QAA instances. Finally, we generate MIS instances of up to 1000 qubits, providing scalable benchmarks for evaluating future, larger processors as they become available. |
| 10 | Breaking the Visual Shortcuts in Multimodal Knowledge-Based Visual Question Answering | 2025-11-28 | 展开Existing Multimodal Knowledge-Based Visual Question Answering (MKB-VQA) benchmarks suffer from "visual shortcuts", as the query image typically matches the primary subject entity of the target document. We demonstrate that models can exploit these shortcuts, achieving comparable results using visual cues alone. To address this, we introduce Relational Entity Text-Image kNowledge Augmented (RETINA) benchmark, automatically constructed using an LLM-driven pipeline, consisting of 120k training and 2k human-curated test set. RETINA contains queries referencing secondary subjects (i.e. related entities) and pairs them with images of these related entities, removing the visual shortcut. When evaluated on RETINA existing models show significantly degraded performance, confirming their reliance on the shortcut. Furthermore, we propose Multi-Image MultImodal Retriever (MIMIR), which enriches document embeddings by augmenting images of multiple related entities, effectively handling RETINA, unlike prior work that uses only a single image per document. Our experiments validate the limitations of existing benchmarks and demonstrate the effectiveness of RETINA and MIMIR. Our project is available at: Project Page. |
| 11 | Can Synthetic Data Improve Symbolic Regression Extrapolation Performance? | 2025-11-27 | 展开Many machine learning models perform well when making predictions within the training data range, but often struggle when required to extrapolate beyond it. Symbolic regression (SR) using genetic programming (GP) can generate flexible models but is prone to unreliable behaviour in extrapolation. This paper investigates whether adding synthetic data can help improve performance in such cases. We apply Kernel Density Estimation (KDE) to identify regions in the input space where the training data is sparse. Synthetic data is then generated in those regions using a knowledge distillation approach: a teacher model generates predictions on new input points, which are then used to train a student model. We evaluate this method across six benchmark datasets, using neural networks (NN), random forests (RF), and GP both as teacher models (to generate synthetic data) and as student models (trained on the augmented data). Results show that GP models can often improve when trained on synthetic data, especially in extrapolation areas. However, the improvement depends on the dataset and teacher model used. The most important improvements are observed when synthetic data from GPe is used to train GPp in extrapolation regions. Changes in interpolation areas show only slight changes. We also observe heterogeneous errors, where model performance varies across different regions of the input space. Overall, this approach offers a practical solution for better extrapolation. Note: An earlier version of this work appeared in the GECCO 2025 Workshop on Symbolic Regression. This arXiv version corrects several parts of the original submission. |
| 12 | All Centers Are at most a Few Tokens Apart: Knowledge Distillation with Domain Invariant Prompt Tuning | 2025-11-27 | 展开Domain generalization is critical in computational pathology (CPath) due to inherent domain shifts caused by variations in staining protocols, scanner devices, and imaging settings across clinical centers. Vision-language models (VLMs), such as PLIP-a pathology-tuned CLIP-trained on image-text pairs across diverse domains, serve as strong knowledge distillation sources. However, their zero-shot performance with predefined prompts remains limited due to sensitivity to prompt variations. Moreover, unlike natural images, histopathology centers lack semantic descriptors (e.g., 'sketch'), making it difficult to define domain-specific prompts for clinical centers. This requires a data-driven approach for learning domain-specific and ultimately class-generic continuous prompts. We propose Domain Invariant Prompt Tuning (DIPT) for knowledge distillation process, a novel step that learns multiple input tokens for each domain. These tokens are trained separately for each domain and are averaged across domains, leading to domain-invariant prompts. Our student model then distills knowledge from PLIP's text encoder by leveraging the prompts learned by DIPT. This leads to alignment of visual features with domain-invariant embeddings, enhancing generalization by training on multiple domains. Our method adds a significant improvement in average F1-score to existing state-of-the-art (SOTA) knowledge distillation approaches in domain generalization with histopathology datasets. This work helps the way of deploying robust CPath models in real-world clinical problems with heterogeneous data sources. |
| 13 | CacheTrap: Injecting Trojans in LLMs without Leaving any Traces in Inputs or Weights | 2025-11-27 | 展开Adversarial weight perturbation has emerged as a concerning threat to LLMs that either use training privileges or system-level access to inject adversarial corruption in model weights. With the emergence of innovative defensive solutions that place system- and algorithm-level checks and corrections in the input and weight spaces, these perturbations are increasingly susceptible to defenses. This work develops a novel perspective on Trojan attacks that generates an attacker-designed model output while leaving no attack traces on the inputs or weights. Such an attack space can be unlocked through corruption of the key-value (KV) cache. In this paper, we introduce CacheTrap, a novel Trojan attack that corrupts the value vectors stored in the KV cache. These vectors capture the dynamic activations for specific token positions and therefore constitute a natural surface for transient, inference-time trigger insertion. The transient nature of these KV values and their dependence on victim input imply additional constraints on our attack, such as a lack of knowledge of the victim's data or domain application, and, consequently, a lack of gradient information. The objective of the proposed CacheTrap is to develop a vulnerable KV bit-searching algorithm so that, once the attack employs the identified bit-flip as a trigger, the model generates targeted behavior, e.g., classifying inputs towards the target class. Moreover, CacheTrap is a data- and gradient-free attack which also has no impact on the model's utility. Our evaluation demonstrates that the proposed attack enables the first successful Trojan attack on LLMs with a single bit flip in the KV cache. In addition, the data-independent nature of the attack ensures that once the attacker identifies the vulnerable bit index, the location remains constant and can be transferred to a wide range of victim tasks/datasets/queries with no overhead. |
| 14 | An Efficient Embedding Based Ad Retrieval with GPU-Powered Feature Interaction | 2025-11-27 | 展开In large-scale advertising recommendation systems, retrieval serves as a critical component, aiming to efficiently select a subset of candidate ads relevant to user behaviors from a massive ad inventory for subsequent ranking and recommendation. The Embedding-Based Retrieval (EBR) methods modeled by the dual-tower network are widely used in the industry to maintain both retrieval efficiency and accuracy. However, the dual-tower model has significant limitations: the embeddings of users and ads interact only at the final inner product computation, resulting in insufficient feature interaction capabilities. Although DNN-based models with both user and ad as input features, allowing for early-stage interaction between these features, are introduced in the ranking stage to mitigate this issue, they are computationally infeasible for the retrieval stage. To bridge this gap, this paper proposes an efficient GPU-based feature interaction for the dual-tower network to significantly improve retrieval accuracy while substantially reducing computational costs. Specifically, we introduce a novel compressed inverted list designed for GPU acceleration, enabling efficient feature interaction computation at scale. To the best of our knowledge, this is the first framework in the industry to successfully implement Wide and Deep in a retrieval system. We apply this model to the real-world business scenarios in Tencent Advertising, and experimental results demonstrate that our method outperforms existing approaches in offline evaluation and has been successfully deployed to Tencent's advertising recommendation system, delivering significant online performance gains. This improvement not only validates the effectiveness of the proposed method, but also provides new practical guidance for optimizing large-scale ad retrieval systems. |
| 15 | Fin3R: Fine-tuning Feed-forward 3D Reconstruction Models via Monocular Knowledge Distillation | 2025-11-27 | 展开We present Fin3R, a simple, effective, and general fine-tuning method for feed-forward 3D reconstruction models. The family of feed-forward reconstruction model regresses pointmap of all input images to a reference frame coordinate system, along with other auxiliary outputs, in a single forward pass. However, we find that current models struggle with fine geometry and robustness due to (\textit{i}) the scarcity of high-fidelity depth and pose supervision and (\textit{ii}) the inherent geometric misalignment from multi-view pointmap regression. Fin3R jointly tackles two issues with an extra lightweight fine-tuning step. We freeze the decoder, which handles view matching, and fine-tune only the image encoder-the component dedicated to feature extraction. The encoder is enriched with fine geometric details distilled from a strong monocular teacher model on large, unlabeled datasets, using a custom, lightweight LoRA adapter. We validate our method on a wide range of models, including DUSt3R, MASt3R, CUT3R, and VGGT. The fine-tuned models consistently deliver sharper boundaries, recover complex structures, and achieve higher geometric accuracy in both single- and multi-view settings, while adding only the tiny LoRA weights, which leave test-time memory and latency virtually unchanged. Project page: \href{http://visual-ai.github.io/fin3r}{https://visual-ai.github.io/fin3r} |
| 16 | Joint Estimation of Sea State and Vessel Parameters Using a Mass-Spring-Damper Equivalence Model | 2025-11-27 | 展开Real-time sea state estimation is vital for applications like shipbuilding and maritime safety. Traditional methods rely on accurate wave-vessel transfer functions to estimate wave spectra from onboard sensors. In contrast, our approach jointly estimates sea state and vessel parameters without needing prior transfer function knowledge, which may be unavailable or variable. We model the wave-vessel system using pseudo mass-spring-dampers and develop a dynamic model for the system. This method allows for recursive modeling of wave excitation as a time-varying input, relaxing prior works' assumption of a constant input. We derive statistically consistent process noise covariance and implement a square root cubature Kalman filter for sensor data fusion. Further, we derive the Posterior Cramer-Rao lower bound to evaluate estimator performance. Extensive Monte Carlo simulations and data from a high-fidelity validated simulator confirm that the estimated wave spectrum matches methods assuming complete transfer function knowledge. |
| 17 | Continual Error Correction on Low-Resource Devices | 2025-11-26 | 展开The proliferation of AI models in everyday devices has highlighted a critical challenge: prediction errors that degrade user experience. While existing solutions focus on error detection, they rarely provide efficient correction mechanisms, especially for resource-constrained devices. We present a novel system enabling users to correct AI misclassifications through few-shot learning, requiring minimal computational resources and storage. Our approach combines server-side foundation model training with on-device prototype-based classification, enabling efficient error correction through prototype updates rather than model retraining. The system consists of two key components: (1) a server-side pipeline that leverages knowledge distillation to transfer robust feature representations from foundation models to device-compatible architectures, and (2) a device-side mechanism that enables ultra-efficient error correction through prototype adaptation. We demonstrate our system's effectiveness on both image classification and object detection tasks, achieving over 50% error correction in one-shot scenarios on Food-101 and Flowers-102 datasets while maintaining minimal forgetting (less than 0.02%) and negligible computational overhead. Our implementation, validated through an Android demonstration app, proves the system's practicality in real-world scenarios. |
| 18 | HarmonicAttack: An Adaptive Cross-Domain Audio Watermark Removal | 2025-11-26 | 展开The availability of high-quality, AI-generated audio raises security challenges such as misinformation campaigns and voice-cloning fraud. A key defense against the misuse of AI-generated audio is by watermarking it, so that it can be easily distinguished from genuine audio. As those seeking to misuse AI-generated audio may thus seek to remove audio watermarks, studying effective watermark removal techniques is critical to being able to objectively evaluate the robustness of audio watermarks against removal. Previous watermark removal schemes either assume impractical knowledge of the watermarks they are designed to remove or are computationally expensive, potentially generating a false sense of confidence in current watermark schemes. We introduce HarmonicAttack, an efficient audio watermark removal method that only requires the basic ability to generate the watermarks from the targeted scheme and nothing else. With this, we are able to train a general watermark removal model that is able to remove the watermarks generated by the targeted scheme from any watermarked audio sample. HarmonicAttack employs a dual-path convolutional autoencoder that operates in both temporal and frequency domains, along with GAN-style training, to separate the watermark from the original audio. When evaluated against state-of-the-art watermark schemes AudioSeal, WavMark, and Silentcipher, HarmonicAttack demonstrates greater watermark removal ability than previous watermark removal methods with near real-time performance. Moreover, while HarmonicAttack requires training, we find that it is able to transfer to out-of-distribution samples with minimal degradation in performance. |
| 19 | Multimodal Robust Prompt Distillation for 3D Point Cloud Models | 2025-11-26 | 展开Adversarial attacks pose a significant threat to learning-based 3D point cloud models, critically undermining their reliability in security-sensitive applications. Existing defense methods often suffer from (1) high computational overhead and (2) poor generalization ability across diverse attack types. To bridge these gaps, we propose a novel yet efficient teacher-student framework, namely Multimodal Robust Prompt Distillation (MRPD) for distilling robust 3D point cloud model. It learns lightweight prompts by aligning student point cloud model's features with robust embeddings from three distinct teachers: a vision model processing depth projections, a high-performance 3D model, and a text encoder. To ensure a reliable knowledge transfer, this distillation is guided by a confidence-gated mechanism which dynamically balances the contribution of all input modalities. Notably, since the distillation is all during the training stage, there is no additional computational cost at inference. Extensive experiments demonstrate that MRPD substantially outperforms state-of-the-art defense methods against a wide range of white-box and black-box attacks, while even achieving better performance on clean data. Our work presents a new, practical paradigm for building robust 3D vision systems by efficiently harnessing multimodal knowledge. |
| 20 | From Static Pathways to Dynamic Mechanisms: A Committor-Based Data-Driven Approach to Chemical Reactions | 2025-11-26 | 展开As computational chemistry methods evolve, dynamic effects have been increasingly recognized to govern chemical reaction pathways in both organic and inorganic systems. Here, we introduce a committor-based workflow that integrates a path-committor-consistent artificial neural network (PCCANN) with an iteratively trained hybrid-DFT-level message passing atomic convolutional encoder (MACE) potential. Beginning with a static nudged elastic band path, PCCANN extracts a committor-consistent string to represent the reactive ensemble. We illustrate the power of this methodology through two representative applications. First, we investigate an SNAr reaction using MACE trained at hybrid DFT level with implicit solvent. The mechanism is found to be concerted, and the dynamic approach reveals a lower barrier than static treatments. Second, we apply the same protocol to the isomerization of protonated isobutanol to protonated 2-butanol, yielding a quantitatively accurate free-energy landscape. We uncover three competing channels: the established concerted mechanism and two asynchronous stepwise routes mediated by water and methyl transfer, all with comparable activation barriers. Notably, the stepwise pathways traverse metastable intermediates that, to the best of our knowledge, have not been described in prior mechanistic studies. Calculated barrier heights and intermediate stabilities are in close agreement with high-level DFT benchmarks, demonstrating the framework's accuracy. Together, these studies highlight mechanistic diversity across distinct systems and establish the synergistic PCCANN-MACE protocol as a proof-of-concept approach for committor-based discovery of complex reaction dynamics. |
| 21 | Actionable and diverse counterfactual explanations incorporating domain knowledge and causal constraints | 2025-11-25 | 展开Counterfactual explanations enhance the actionable interpretability of machine learning models by identifying the minimal changes required to achieve a desired outcome of the model. However, existing methods often ignore the complex dependencies in real-world datasets, leading to unrealistic or impractical modifications. Motivated by cybersecurity applications in the email marketing domain, we propose a method for generating Diverse, Actionable, and kNowledge-Constrained Explanations (DANCE), which incorporates feature dependencies and causal constraints to ensure plausibility and real-world feasibility of counterfactuals. Our method learns linear and nonlinear constraints from data or integrates expert-provided dependency graphs, ensuring counterfactuals are plausible and actionable. By maintaining consistency with feature relationships, the method produces explanations that align with real-world constraints. Additionally, it balances plausibility, diversity, and sparsity, effectively addressing key limitations in existing algorithms. The work is developed based on a real-life case study with Freshmail, the largest email marketing company in Poland and supported by a joint R&D project Sendguard. Furthermore, we provide an extensive evaluation using 140 public datasets, which highlights its ability to generate meaningful, domain-relevant counterfactuals that outperform other existing approaches based on widely used metrics. The source code for reproduction of the results can be found in a GitHub repository we provide. |
| 22 | LLMAID: Identifying AI Capabilities in Android Apps with LLMs | 2025-11-24 | 展开Recent advancements in artificial intelligence (AI) and its widespread integration into mobile software applications have received significant attention, highlighting the growing prominence of AI capabilities in modern software systems. However, the inherent hallucination and reliability issues of AI continue to raise persistent concerns. Consequently, application users and regulators increasingly ask critical questions such as: Does the application incorporate AI capabilities? and What specific types of AI functionalities are embedded? Preliminary efforts have been made to identify AI capabilities in mobile software; however, existing approaches mainly rely on manual inspection and rule-based heuristics. These methods are not only costly and time-consuming but also struggle to adapt advanced AI techniques. To address the limitations of existing methods, we propose LLMAID (Large Language Model for AI Discovery). LLMAID includes four main tasks: (1) candidate extraction, (2) knowledge base interaction, (3) AI capability analysis and detection, and (4) AI service summarization. We apply LLMAID to a dataset of 4,201 Android applications and demonstrate that it identifies 242% more real-world AI apps than state-of-the-art rule-based approaches. Our experiments show that LLM4AID achieves high precision and recall, both exceeding 90%, in detecting AI-related components. Additionally, a user study indicates that developers find the AI service summaries generated by LLMAID to be more informative and preferable to the original app descriptions. Finally, we leverage LLMAID to perform an empirical analysis of AI capabilities across Android apps. The results reveal a strong concentration of AI functionality in computer vision (54.80%), with object detection emerging as the most common task (25.19%). |
| 23 | Toward Honest Language Models for Deductive Reasoning | 2025-11-12 | 展开Deductive reasoning is the process of deriving conclusions strictly from the given premises, without relying on external knowledge. We define honesty in this setting as a model's ability to respond only when the conclusion is logically entailed by the premises, and to abstain otherwise. However, current language models often fail to reason honestly, producing unwarranted answers when the input is insufficient. To study this challenge, we formulate honest deductive reasoning as multi-step tasks where models must either derive the correct conclusion or abstain. We curate two datasets from graph structures, one for linear algebra and one for logical inference, and introduce unanswerable cases by randomly perturbing an edge in half of the instances. We find that prompting and existing training methods, including GRPO with or without supervised fine-tuning initialization, struggle on these tasks. In particular, GRPO optimize only for final task outcomes, leaving models vulnerable to collapse when negative rewards dominate early training. To address this, we propose ACNCHOR, a reinforcement learning method that injects ground truth trajectories into rollouts, preventing early training collapse. Our results demonstrate that this method stabilizes learning and significantly improves the overall reasoning performance, underscoring the importance of training dynamics for enabling honest deductive reasoning in language models. |
| 24 | Continual Learning of Domain Knowledge from Human Feedback in Text-to-SQL | 2025-11-10 | 展开Large Language Models (LLMs) can generate SQL queries from natural language questions but struggle with database-specific schemas and tacit domain knowledge. We introduce a framework for continual learning from human feedback in text-to-SQL, where a learning agent receives natural language feedback to refine queries and distills the revealed knowledge for reuse on future tasks. This distilled knowledge is stored in a structured memory, enabling the agent to improve execution accuracy over time. We design and evaluate multiple variations of a learning agent architecture that vary in how they capture and retrieve past experiences. Experiments on the BIRD benchmark Dev set show that memory-augmented agents, particularly the Procedural Agent, achieve significant accuracy gains and error reduction by leveraging human-in-the-loop feedback. Our results highlight the importance of transforming tacit human expertise into reusable knowledge, paving the way for more adaptive, domain-aware text-to-SQL systems that continually learn from a human-in-the-loop. |
| 25 | Experimental overview on the charmed baryon decays | 2025-09-23 | 展开The charmed baryon was first observed experimentally in 1975, one year after the charm quark's confirmation via the discovery of the |
| 26 | RELATE: Relation Extraction in Biomedical Abstracts with LLMs and Ontology Constraints | 2025-09-23 | 展开Biomedical knowledge graphs (KGs) are vital for drug discovery and clinical decision support but remain incomplete. Large language models (LLMs) excel at extracting biomedical relations, yet their outputs lack standardization and alignment with ontologies, limiting KG integration. We introduce RELATE, a three-stage pipeline that maps LLM-extracted relations to standardized ontology predicates using ChemProt and the Biolink Model. The pipeline includes: (1) ontology preprocessing with predicate embeddings, (2) similarity-based retrieval enhanced with SapBERT, and (3) LLM-based reranking with explicit negation handling. This approach transforms relation extraction from free-text outputs to structured, ontology-constrained representations. On the ChemProt benchmark, RELATE achieves 52% exact match and 94% accuracy@10, and in 2,400 HEAL Project abstracts, it effectively rejects irrelevant associations (0.4%) and identifies negated assertions. RELATE captures nuanced biomedical relationships while ensuring quality for KG augmentation. By combining vector search with contextual LLM reasoning, RELATE provides a scalable, semantically accurate framework for converting unstructured biomedical literature into standardized KGs. |
| 27 | SegDINO3D: 3D Instance Segmentation Empowered by Both Image-Level and Object-Level 2D Features | 2025-09-19 | 展开In this paper, we present SegDINO3D, a novel Transformer encoder-decoder framework for 3D instance segmentation. As 3D training data is generally not as sufficient as 2D training images, SegDINO3D is designed to fully leverage 2D representation from a pre-trained 2D detection model, including both image-level and object-level features, for improving 3D representation. SegDINO3D takes both a point cloud and its associated 2D images as input. In the encoder stage, it first enriches each 3D point by retrieving 2D image features from its corresponding image views and then leverages a 3D encoder for 3D context fusion. In the decoder stage, it formulates 3D object queries as 3D anchor boxes and performs cross-attention from 3D queries to 2D object queries obtained from 2D images using the 2D detection model. These 2D object queries serve as a compact object-level representation of 2D images, effectively avoiding the challenge of keeping thousands of image feature maps in the memory while faithfully preserving the knowledge of the pre-trained 2D model. The introducing of 3D box queries also enables the model to modulate cross-attention using the predicted boxes for more precise querying. SegDINO3D achieves the state-of-the-art performance on the ScanNetV2 and ScanNet200 3D instance segmentation benchmarks. Notably, on the challenging ScanNet200 dataset, SegDINO3D significantly outperforms prior methods by +8.6 and +6.8 mAP on the validation and hidden test sets, respectively, demonstrating its superiority. |
| 28 | Privacy-Preserving Reasoning with Knowledge-Distilled Parametric Retrieval Augmented Generation | 2025-09-01 | 展开The current RAG system requires uploading plaintext documents to the cloud, risking private data leakage. Parametric RAG (PRAG) encodes documents as LoRA parameters within LLMs, offering a possible way to reduce exposure of raw content. However, it still faces two issues: (1) PRAG demands synthesizing QA pairs and fine-tuning LLM for each individual document to create its corresponding LoRA, leading to unacceptable inference latency. (2) The performance of PRAG relies solely on synthetic QA data while lacking internal alignment with standard RAG, resulting in poor generalization on out-of-distribution(OOD) inputs. Therefore, achieving high-efficiency parameterization while maintaining RAG-level performance remains a critical challenge for privacy-preserving reasoning. In this paper, we propose DistilledPRAG, a generalizable knowledge-distilled parametric RAG model aligned with standard RAG in document structure and parameter activation. We first synthesize QA pairs from single and multi-documents to enhance cross-document reasoning. Then, we mask the plaintext documents with a special token and translate them to LoRA via a parameter generator, maintaining the standard RAG document structure. Finally, guided by synthetic QA data, we train the parameter generator to match standard RAG's hidden states and output logits, enabling RAG-style reasoning without original documents. Experiments on four QA datasets show that DistilledPRAG outperforms baselines in accuracy and generalizes well on OOD data. |
| 29 | Dual-Model Weight Selection and Self-Knowledge Distillation for Medical Image Classification | 2025-08-28 | 展开We propose a novel medical image classification method that integrates dual-model weight selection with self-knowledge distillation (SKD). In real-world medical settings, deploying large-scale models is often limited by computational resource constraints, which pose significant challenges for their practical implementation. Thus, developing lightweight models that achieve comparable performance to large-scale models while maintaining computational efficiency is crucial. To address this, we employ a dual-model weight selection strategy that initializes two lightweight models with weights derived from a large pretrained model, enabling effective knowledge transfer. Next, SKD is applied to these selected models, allowing the use of a broad range of initial weight configurations without imposing additional excessive computational cost, followed by fine-tuning for the target classification tasks. By combining dual-model weight selection with self-knowledge distillation, our method overcomes the limitations of conventional approaches, which often fail to retain critical information in compact models. Extensive experiments on publicly available datasets-chest X-ray images, lung computed tomography scans, and brain magnetic resonance imaging scans-demonstrate the superior performance and robustness of our approach compared to existing methods. |
| 30 | CAMA: Enhancing Mathematical Reasoning in Large Language Models with Causal Knowledge | 2025-08-04 | 展开Large Language Models (LLMs) have demonstrated strong performance across a wide range of tasks, yet they still struggle with complex mathematical reasoning, a challenge fundamentally rooted in deep structural dependencies. To address this challenge, we propose \textbf{CA}usal \textbf{MA}thematician (\textbf{CAMA}), a two-stage causal framework that equips LLMs with explicit, reusable mathematical structure. In the learning stage, CAMA first constructs the \textbf{M}athematical \textbf{C}ausal \textbf{G}raph (\textbf{MCG}), a high-level representation of solution strategies, by combining LLM priors with causal discovery algorithms applied to a corpus of question-solution pairs. The resulting MCG encodes essential knowledge points and their causal dependencies. To better align the graph with downstream reasoning tasks, CAMA further refines the MCG through iterative feedback derived from a selected subset of the question-solution pairs. In the reasoning stage, given a new question, CAMA dynamically extracts a task-relevant subgraph from the MCG, conditioned on both the question content and the LLM's intermediate reasoning trace. This subgraph, which encodes the most pertinent knowledge points and their causal dependencies, is then injected back into the LLM to guide its reasoning process. Empirical results on real-world datasets show that CAMA significantly improves LLM performance on challenging mathematical problems. Furthermore, our experiments demonstrate that structured guidance consistently outperforms unstructured alternatives, and that incorporating asymmetric causal relationships yields greater improvements than using symmetric associations alone. |
| 31 | Predicting Video Slot Attention Queries from Random Slot-Feature Pairs | 2025-08-02 | 展开Unsupervised video Object-Centric Learning (OCL) is promising as it enables object-level scene representation and understanding as we humans do. Mainstream video OCL methods adopt a recurrent architecture: An aggregator aggregates current video frame into object features, termed slots, under some queries; A transitioner transits current slots to queries for the next frame. This is an effective architecture but all existing implementations both (\textit{i1}) neglect to incorporate next frame features, the most informative source for query prediction, and (\textit{i2}) fail to learn transition dynamics, the knowledge essential for query prediction. To address these issues, we propose Random Slot-Feature pair for learning Query prediction (RandSF.Q): (\textit{t1}) We design a new transitioner to incorporate both slots and features, which provides more information for query prediction; (\textit{t2}) We train the transitioner to predict queries from slot-feature pairs randomly sampled from available recurrences, which drives it to learn transition dynamics. Experiments on scene representation demonstrate that our method surpass existing video OCL methods significantly, e.g., up to 10 points on object discovery, setting new state-of-the-art. Such superiority also benefits downstream tasks like scene understanding. Source Code, Model Checkpoints, Training Logs: https://github.com/Genera1Z/RandSF.Q |
| 32 | Gradient-Based Program Repair: Fixing Bugs in Continuous Program Spaces | 2025-05-23 | 展开Automatic program repair seeks to generate correct code from buggy programs, with most approaches searching the correct program in a discrete, symbolic space of source code tokens. This symbolic search is fundamentally limited by its inability to directly reason about program behavior. We introduce Gradient-Based Program Repair (GBPR), a new paradigm that reframes program repair as continuous optimization in a differentiable numerical program space. Our core insight is to compile symbolic programs into differentiable numerical representations, enabling search in the numerical program space directly guided by program behavior. To evaluate GBPR, we present RaspBugs, a new benchmark of 1,466 buggy symbolic RASP programs and their respective numerical representations. Our experiments demonstrate that GBPR can effectively repair buggy symbolic programs by gradient-based optimization in the numerical program space, with convincing repair trajectories. To our knowledge, we are the first to state program repair as continuous optimization in a numerical program space. Our work establishes a new direction for program repair research, bridging two rich worlds: continuous optimization and program behavior. |
| 33 | Anomaly Resilient Temporal QoS Prediction using Hypergraph Convoluted Transformer Network | 2024-10-23 | 展开Quality-of-Service (QoS) prediction is a critical task in the service lifecycle, enabling precise and adaptive service recommendations by anticipating performance variations over time in response to evolving network uncertainties and user preferences. However, contemporary QoS prediction methods frequently encounter data sparsity and cold-start issues, which hinder accurate QoS predictions and limit the ability to capture diverse user preferences. Additionally, these methods often assume QoS data reliability, neglecting potential credibility issues such as outliers and the presence of greysheep users and services with atypical invocation patterns. Furthermore, traditional approaches fail to leverage diverse features, including domain-specific knowledge and complex higher-order patterns, essential for accurate QoS predictions. In this paper, we introduce a real-time, trust-aware framework for temporal QoS prediction to address the aforementioned challenges, featuring an end-to-end deep architecture called the Hypergraph Convoluted Transformer Network (HCTN). HCTN combines a hypergraph structure with graph convolution over hyper-edges to effectively address high-sparsity issues by capturing complex, high-order correlations. Complementing this, the transformer network utilizes multi-head attention along with parallel 1D convolutional layers and fully connected dense blocks to capture both fine-grained and coarse-grained dynamic patterns. Additionally, our approach includes a sparsity-resilient solution for detecting greysheep users and services, incorporating their unique characteristics to improve prediction accuracy. Trained with a robust loss function resistant to outliers, HCTN demonstrated state-of-the-art performance on the large-scale WSDREAM-2 datasets for response time and throughput. |
| 34 | Many-body quantum resources of graph states | 2024-10-16 | 展开Characterizing the non-classical correlations of a complex many-body system is an important part of quantum technologies. A versatile tool for such a task is one that scales well with the size of the system and which can be both easily computed and measured. In this work we focus on graph states, that are promising platforms for quantum computation, simulation and metrology. We consider four topologies, namely the star graph states with edges, Turán graphs, |
| 35 | A Gray-box Attack against Latent Diffusion Model-based Image Editing by Posterior Collapse | 2024-08-20 | 展开Recent advancements in Latent Diffusion Models (LDMs) have revolutionized image synthesis and manipulation, raising significant concerns about data misappropriation and intellectual property infringement. While adversarial attacks have been extensively explored as a protective measure against such misuse of generative AI, current approaches are severely limited by their heavy reliance on model-specific knowledge and substantial computational costs. Drawing inspiration from the posterior collapse phenomenon observed in VAE training, we propose the Posterior Collapse Attack (PCA), a novel framework for protecting images from unauthorized manipulation. Through comprehensive theoretical analysis and empirical validation, we identify two distinct collapse phenomena during VAE inference: diffusion collapse and concentration collapse. Based on this discovery, we design a unified loss function that can flexibly achieve both types of collapse through parameter adjustment, each corresponding to different protection objectives in preventing image manipulation. Our method significantly reduces dependence on model-specific knowledge by requiring access to only the VAE encoder, which constitutes less than 4% of LDM parameters. Notably, PCA achieves prompt-invariant protection by operating on the VAE encoder before text conditioning occurs, eliminating the need for empty prompt optimization required by existing methods. This minimal requirement enables PCA to maintain adequate transferability across various VAE-based LDM architectures while effectively preventing unauthorized image editing. Extensive experiments show PCA outperforms existing techniques in protection effectiveness, computational efficiency (runtime and VRAM), and generalization across VAE-based LDM variants. Our code is available at https://github.com/ZhongliangGuo/PosteriorCollapseAttack. |
| 36 | Learning Rules from Rewards | 2022-03-25 | 展开Humans can flexibly generalize knowledge across domains by leveraging structured relational representations. While prior research has shown how such representations support analogical reasoning, less is known about how they are recruited to guide adaptive behavior. We address this gap by introducing the Relational Regression Tree Learner (RRTL), a model that incrementally builds policies over structured relational inputs by selecting task-relevant relations during the learning process. RRTL is grounded in the framework of relational reinforcement learning but diverges from traditional approaches by focusing on ground (i.e., non-variabilized) rules that refer to specific object configurations. Across three Atari games of increasing relational complexity (Breakout, Pong, Demon Attack), the model learns to act effectively by identifying a small set of relevant relations from a broad pool of candidate relations. A comparative version of the model, which partitions the state space using relative magnitude values (e.g., "more", "same", "less"), showed more robust learning than a version using logical (binary) splits. These results provide a proof of principle that reinforcement signals can guide the selection of structured representations, offering a computational framework for understanding how relational knowledge is learned and deployed in adaptive behavior. |
| 序号 | 标题 | 日期 | 摘要 |
|---|---|---|---|
| 1 | Arctic Auctions, Linear Fisher Markets, and Rational Convex Programs | 2025-11-26 | 展开This paper unifies two foundational constructs from economics and algorithmic game theory, the Arctic Auction and the linear Fisher market, to address the efficient allocation of differentiated goods in complex markets. Our main contributions are showing that an equilibrium for the Arctic Auction is captured by a Rational Convex Program, and deriving the first combinatorial polynomial-time algorithm for computing Arctic Auction equilibria. |
| 2 | Impartial Games with Activeness | 2025-11-26 | 展开A combinatorial game is a two-player game without hidden information or chance elements. The main object of combinatorial game theory is to obtain the outcome, which player has a winning strategy, of a given combinatorial game. Positions of many well-known combinatorial games are naturally decomposed into a disjunctive sum of multiple components and can be analyzed independently for each component. Therefore, the study of disjunctive sums is a major topic in combinatorial game theory. Combinatorial games in which both players have the same set of possible moves for every position are called impartial games. In the normal-play convention, it is known that the outcome of a disjunctive sum of impartial games can be obtained by computing the Grundy number of each term. The theory of impartial games is generalized in various forms. This paper proposes another generalization of impartial games to a new framework, impartial games with activeness: each game is assigned a status of either |
| 3 | $\mathcal{L}\mathcal{R}$-Ending partisan rulesets | 2025-11-18 | 展开In this paper, we consider |
| 4 | Finding Kissing Numbers with Game-theoretic Reinforcement Learning | 2025-11-17 | 展开Since Isaac Newton first studied the Kissing Number Problem in 1694, determining the maximal number of non-overlapping spheres around a central sphere has remained a fundamental challenge. This problem represents the local analogue of Hilbert's 18th problem on sphere packing, bridging geometry, number theory, and information theory. Although significant progress has been made through lattices and codes, the irregularities of high-dimensional geometry and exponentially growing combinatorial complexity beyond 8 dimensions, which exceeds the complexity of Go game, limit the scalability of existing methods. Here we model this problem as a two-player matrix completion game and train the game-theoretic reinforcement learning system, PackingStar, to efficiently explore high-dimensional spaces. The matrix entries represent pairwise cosines of sphere center vectors; one player fills entries while another corrects suboptimal ones, jointly maximizing the matrix size, corresponding to the kissing number. This cooperative dynamics substantially improves sample quality, making the extremely large spaces tractable. PackingStar reproduces previous configurations and surpasses all human-known records from dimensions 25 to 31, with the configuration in 25 dimensions geometrically corresponding to the Leech lattice and suggesting possible optimality. It achieves the first breakthrough beyond rational structures from 1971 in 13 dimensions and discovers over 6000 new structures in 14 and other dimensions. These results demonstrate AI's power to explore high-dimensional spaces beyond human intuition and open new pathways for the Kissing Number Problem and broader geometry problems. |
| 5 | Visibility-aware Satellite Selection and Resource Allocation in Multi-Orbit LEO Networks | 2025-11-16 | 展开Multi orbit low earth orbit (LEO) satellites communication is envisioned as a key infrastructure to deliver global coverage, enabling future services from space air ground integrated networks.However, the optimized design of LEO which jointly addresses satellite selection, association control, and resource scheduling while accounting for dynamic visibility in multi orbit constellations still remains open. Satellites moving along distinct orbital planes yield phase shifted ground tracks and heterogeneous, time varying coverage patterns that significantly complicate the optimization.To bridge the gap, we propose a dynamic visibility aware multi orbit satellite selection framework which can determine the optimal serving satellites across orbital layers. The framework is built upon Markov approximation and matching game theory. Specifically, we formulate a combinatorial optimization problem that maximizes the sum rate under per satellite power budgets. The problem is NP hard , combining discrete user association (UA) decisions with continuous power allocation, and an inherently non convex sum rate maximization objective. We address it through a problem specific Markov approximation. Moreover, we alternately solve UA or bandwidth allocation via a matching game and power allocation via a Lagrangian dual program, which together form a block coordinate descent method tailored to this problem. Simulation results show that the proposed algorithm converges to a suboptimal solution across all scenarios. Extensive experiments against four state of the art baselines further demonstrate that our algorithm achieves, on average, approximately 7.85% higher sum rate than the best performing baseline. |
| 6 | A Complete Classification of Ideal Chomp Games on Low-Rank Algebras | 2025-11-03 | 展开We completely classify winning strategies in the Ideal Chomp Game played on |
| 7 | Self interest cumulative subtraction games | 2025-10-28 | 展开Subtraction games have a rich literature as normal-play combinatorial games (e.g., Berlekamp, Conway, and Guy, 1982). Recently, the theory has been extended to zero-sum scoring play (Cohensius et al. 2019). Here, we take the approach of cumulative self-interest games, as introduced in a recent framework preprint by Larsson, Meir, and Zick. By adapting standard Pure Subgame Perfect Equilibria (PSPE) from classical game theory, players must declare and commit to acting either |
| 8 | Xiangqi-R1: Enhancing Spatial Strategic Reasoning in LLMs for Chinese Chess via Reinforcement Learning | 2025-07-16 | 展开Game playing has long served as a fundamental benchmark for evaluating Artificial General Intelligence. While Large Language Models (LLMs) have demonstrated impressive capabilities in general reasoning, their effectiveness in spatial strategic reasoning, which is critical for complex and fully observable board games, remains insufficiently explored. In this work, we adopt Chinese Chess (Xiangqi) as a challenging and rich testbed due to its intricate rules and spatial complexity. To advance LLMs' strategic competence in such environments, we propose a training framework tailored to Xiangqi, built upon a large-scale dataset of five million board-move pairs enhanced with expert annotations and engine evaluations. Building on this foundation, we introduce Xiangqi-R1, a 7B-parameter model trained in multi-stage manner. Our Experimental results indicate that, despite their size and power, general-purpose LLMs struggle to achieve satisfactory performance in these tasks. Compared to general-purpose LLMs, Xiangqi-R1 greatly advances with an 18% rise in move legality and a 22% boost in analysis accuracy. Our results point to a promising path for creating general strategic intelligence in complex areas. |
| 9 | Deep Reinforcement Learning Xiangqi Player with Monte Carlo Tree Search | 2025-06-18 | 展开This paper presents a Deep Reinforcement Learning (DRL) system for Xiangqi (Chinese Chess) that integrates neural networks with Monte Carlo Tree Search (MCTS) to enable strategic self-play and self-improvement. Addressing the underexplored complexity of Xiangqi, including its unique board layout, piece movement constraints, and victory conditions, our approach combines policy-value networks with MCTS to simulate move consequences and refine decision-making. By overcoming challenges such as Xiangqi's high branching factor and asymmetrical piece dynamics, our work advances AI capabilities in culturally significant strategy games while providing insights for adapting DRL-MCTS frameworks to domain-specific rule systems. |
| 10 | Computational and Algebraic Structure of Board Games | 2025-02-18 | 展开We provide two methodologies in the area of computation theory to solve optimal strategies for board games such as Xi Gua Qi and Go. From experimental results, we find relevance to graph theory, matrix representation, and mathematical consciousness. We prove that the decision strategy of movement for Xi Gua Qi and Chinese checker games belongs to a subset that is neither a ring nor a group over set Y={-1,0,1}. Additionally, the movement for any board game with two players belongs to a subset that is neither a ring nor a group from the razor of Occam. We derive the closed form of the transition matrix for any board game with two players such as chess and Chinese chess. We discover that the element of the transition matrix belongs to a rational number. We propose a different methodology based on algebra theory to analyze the complexity of board games in their entirety, instead of being limited solely to endgame results. It is probable that similar decision processes of people may also belong to a matrix representation that is neither a ring nor a group. |
| 11 | RemoteChess: Enhancing Older Adults' Social Connectedness via Designing a Virtual Reality Chinese Chess (Xiangqi) Community | 2025-02-17 | 展开The decline of social connectedness caused by distance and physical limitations severely affects older adults' well-being and mental health. While virtual reality (VR) is promising for older adults to socialize remotely, existing social VR designs primarily focus on verbal communication (e.g., reminiscent, chat). Actively engaging in shared activities is also an important aspect of social connection. We designed RemoteChess, which constructs a social community and a culturally relevant activity (i.e., Chinese chess) for older adults to play while engaging in social interaction. We conducted a user study with groups of older adults interacting with each other through RemoteChess. Our findings indicate that RemoteChess enhanced participants' social connectedness by offering familiar environments, culturally relevant social catalysts, and asymmetric interactions. We further discussed design guidelines for designing culturally relevant social activities in VR to promote social connectedness for older adults. |
| 12 | Complete Implementation of WXF Chinese Chess Rules | 2024-12-23 | 展开Unlike repetitions in Western Chess where all repetitions are draws, repetitions in Chinese Chess could result in a win, draw, or loss depending on the kind of repetition being made by both players. One of the biggest hurdles facing Chinese Chess application development is a proper system for judging games correctly. This paper introduces a complete algorithm for ruling the WXF rules correctly in all 110 example cases found in the WXF manual. We introduce several novel optimizations for speeding up the repetition handling without compromising the program correctness. This algorithm is usable in engines, and we saw a total increase in playing strength by +10 point rating increase, or an increased 5% winrate when integrating this approach into our prototype engine. |
| 13 | Mastering Chinese Chess AI (Xiangqi) Without Search | 2024-10-07 | 展开We have developed a high-performance Chinese Chess AI that operates without reliance on search algorithms. This AI has demonstrated the capability to compete at a level commensurate with the top 0.1% of human players. By eliminating the search process typically associated with such systems, this AI achieves a Queries Per Second (QPS) rate that exceeds those of systems based on the Monte Carlo Tree Search (MCTS) algorithm by over a thousandfold and surpasses those based on the AlphaBeta pruning algorithm by more than a hundredfold. The AI training system consists of two parts: supervised learning and reinforcement learning. Supervised learning provides an initial human-like Chinese chess AI, while reinforcement learning, based on supervised learning, elevates the strength of the entire AI to a new level. Based on this training system, we carried out enough ablation experiments and discovered that 1. The same parameter amount of Transformer architecture has a higher performance than CNN on Chinese chess; 2. Possible moves of both sides as features can greatly improve the training process; 3. Selective opponent pool, compared to pure self-play training, results in a faster improvement curve and a higher strength limit. 4. Value Estimation with Cutoff(VECT) improves the original PPO algorithm training process and we will give the explanation. |
| 14 | XQSV: A Structurally Variable Network to Imitate Human Play in Xiangqi | 2024-07-05 | 展开In this paper, we introduce an innovative deep learning architecture, termed Xiangqi Structurally Variable (XQSV), designed to emulate the behavioral patterns of human players in Xiangqi, or Chinese Chess. The unique attribute of XQSV is its capacity to alter its structural configuration dynamically, optimizing performance for the task based on the particular subset of data on which it is trained. We have incorporated several design improvements to significantly enhance the network's predictive accuracy, including a local illegal move filter, an Elo range partitioning, a sequential one-dimensional input, and a simulation of imperfect memory capacity. Empirical evaluations reveal that XQSV attains a predictive accuracy of approximately 40%, with its performance peaking within the trained Elo range. This indicates the model's success in mimicking the play behavior of individuals within that specific range. A three-terminal Turing Test was employed to demonstrate that the XQSV model imitates human behavior more accurately than conventional Xiangqi engines, rendering it indistinguishable from actual human opponents. Given the inherent nondeterminism in human gameplay, we propose two supplementary relaxed evaluation metrics. To our knowledge, XQSV represents the first model to mimic Xiangqi players. |
| 15 | Fair Core Imputations for the Assignment Game: New Solution Concepts and Efficient Algorithms | 2024-02-18 | 展开The assignment game is a classical model for profit sharing and a cornerstone of cooperative game theory. While an imputation in its core guarantees fairness among coalitions, it provides no fairness guarantee at the level of individual agents: single agents or one sided coalitions have zero standalone value and may receive arbitrarily small payoffs. Motivated by the growing focus on individual level fairness, we ask: Can one select a core imputation that is also fair to individuals? We introduce three individual fairness driven solution concepts, each promoting equity in a different way. The leximin and leximax core imputations extend max min and min max fairness to uplift the least advantaged and constrain the most advantaged agents, respectively. The min spread core imputation minimizes the gap between the largest and smallest positive payoffs, promoting equitable profit distribution. For all three solution concepts, we develop combinatorial, strongly polynomial algorithms. The leximin and leximax algorithms are based on a novel adaptation of the primal dual paradigm, while the min spread algorithm combines partial executions of the first two. We expect our work to revive innovation on the potent primal dual paradigm as well as promote further work on the algorithmic study of fairness and stability. |
| 16 | Shogi and Frieze group | 2023-11-15 | 展开Shogi is a traditional Japanese strategy board game in the same family as chess, chaturanga, and xiangqi, and has been theoretically studied from various aspects. The research on recommended sequences of moves in each opening of shogi is called joseki; how to use a rook (Static Rook and Ranging Rook), or how to develop a castle, etc. Also, many pieces of tsume shogi, artistic shogi miniature problems, in which the opponent's king is checkmated by a series of checks, have been created involving various beautiful techniques such as "saw" and "puzzle ring". In addition, the rapid development of AI in recent years has led to the pursuit of the best possible moves in shogi. In this paper, we move away from the study of winning and losing in shogi and focus on the mathematical aspects of the movement of shogi pieces. We propose to correspond movements of shogi pieces to a set of geometrical patterns constructed by the shape of shogi pieces and representing the Frieze group through the condition regarding the neighborhood of arrangements of given shogi pieces. Although the discovery of this correspondence does not lead to a winning strategy for shogi, however, it does demonstrate a curious involvement between the traditional Japanese board game and Western mathematics. |
| 17 | JiangJun: Mastering Xiangqi by Tackling Non-Transitivity in Two-Player Zero-Sum Games | 2023-08-09 | 展开This paper presents an empirical exploration of non-transitivity in perfect-information games, specifically focusing on Xiangqi, a traditional Chinese board game comparable in game-tree complexity to chess and shogi. By analyzing over 10,000 records of human Xiangqi play, we highlight the existence of both transitive and non-transitive elements within the game's strategic structure. To address non-transitivity, we introduce the JiangJun algorithm, an innovative combination of Monte-Carlo Tree Search (MCTS) and Policy Space Response Oracles (PSRO) designed to approximate a Nash equilibrium. We evaluate the algorithm empirically using a WeChat mini program and achieve a Master level with a 99.41% win rate against human players. The algorithm's effectiveness in overcoming non-transitivity is confirmed by a plethora of metrics, such as relative population performance and visualization results. Our project site is available at \url{https://sites.google.com/view/jiangjun-site/}. |
| 18 | Niel's Chess -- Rules for Xiangqi | 2023-06-27 | 展开In this paper, the rules of Niel's Chess are adapted to the game of Xiangqi, following the idea that the River and the Palaces play an important role in restricting and enabling chess pieces in their movements. |
| 序号 | 标题 | 日期 | 摘要 |
|---|---|---|---|
| 1 | ThetaEvolve: Test-time Learning on Open Problems | 2025-11-28 | 展开Recent advances in large language models (LLMs) have enabled breakthroughs in mathematical discovery, exemplified by AlphaEvolve, a closed-source system that evolves programs to improve bounds on open problems. However, it relies on ensembles of frontier LLMs to achieve new bounds and is a pure inference system that models cannot internalize the evolving strategies. We introduce ThetaEvolve, an open-source framework that simplifies and extends AlphaEvolve to efficiently scale both in-context learning and Reinforcement Learning (RL) at test time, allowing models to continually learn from their experiences in improving open optimization problems. ThetaEvolve features a single LLM, a large program database for enhanced exploration, batch sampling for higher throughput, lazy penalties to discourage stagnant outputs, and optional reward shaping for stable training signals, etc. ThetaEvolve is the first evolving framework that enable a small open-source model, like DeepSeek-R1-0528-Qwen3-8B, to achieve new best-known bounds on open problems (circle packing and first auto-correlation inequality) mentioned in AlphaEvolve. Besides, across two models and four open tasks, we find that ThetaEvolve with RL at test-time consistently outperforms inference-only baselines, and the model indeed learns evolving capabilities, as the RL-trained checkpoints demonstrate faster progress and better final performance on both trained target task and other unseen tasks. We release our code publicly: https://github.com/ypwang61/ThetaEvolve |
| 2 | Evaluating LLMs for One-Shot Patching of Real and Artificial Vulnerabilities | 2025-11-28 | 展开Automated vulnerability patching is crucial for software security, and recent advancements in Large Language Models (LLMs) present promising capabilities for automating this task. However, existing research has primarily assessed LLMs using publicly disclosed vulnerabilities, leaving their effectiveness on related artificial vulnerabilities largely unexplored. In this study, we empirically evaluate the patching effectiveness and complementarity of several prominent LLMs, such as OpenAI's GPT variants, LLaMA, DeepSeek, and Mistral models, using both real and artificial vulnerabilities. Our evaluation employs Proof-of-Vulnerability (PoV) test execution to concretely assess whether LLM-generated source code successfully patches vulnerabilities. Our results reveal that LLMs patch real vulnerabilities more effectively compared to artificial ones. Additionally, our analysis reveals significant variability across LLMs in terms of overlapping (multiple LLMs patching the same vulnerabilities) and complementarity (vulnerabilities patched exclusively by a single LLM), emphasizing the importance of selecting appropriate LLMs for effective vulnerability patching. |
| 3 | Functional Program Synthesis with Higher-Order Functions and Recursion Schemes | 2025-11-28 | 展开Program synthesis is the process of generating a computer program following a set of specifications, such as a set of input-output examples. It can be modeled as a search problem in which the search space is the set of all valid programs. As the search space is vast, brute force is usually not feasible, and search heuristics, such as genetic programming, also have difficulty navigating it without guidance. This text presents 2 novel GP algorithms that synthesize pure, typed, and functional programs: HOTGP and Origami. HOTGP uses strong types and a functional grammar, synthesizing Haskell code, with support for higher-order functions, |
| 4 | FLIMs: Fault Localization Interference Mutants, Definition, Recognition and Mitigation | 2025-11-28 | 展开Mutation-based Fault Localization (MBFL) has been widely explored for automated software debugging, leveraging artificial mutants to identify faulty code entities. However, MBFL faces significant challenges due to interference mutants generated from non-faulty code entities but can be killed by failing tests. These mutants mimic the test sensitivity behaviors of real faulty code entities and weaken the effectiveness of fault localization. To address this challenge, we introduce the concept of Fault Localization Interference Mutants (FLIMs) and conduct a theoretical analysis based on the Reachability, Infection, Propagation, and Revealability (RIPR) model, identifying four distinct interference causes. Building on this, we propose a novel approach to semantically recognize and mitigate FLIMs using LLM-based semantic analysis, enhanced by fine-tuning techniques and confidence estimation strategies to address LLM output instability. The recognized FLIMs are then mitigated by refining the suspiciousness scores calculated from MBFL techniques. We integrate FLIM recognition and mitigation into the MBFL workflow, developing MBFL-FLIM, a fault localization framework that enhances MBFL's effectiveness by reducing misleading interference while preserving real fault-revealing information. Our empirical experiments on the Defects4J benchmark with 395 program versions using eight LLMs demonstrate MBFL-FLIM's superiority over traditional SBFL and MBFL methods, advanced dynamic feature-based approaches, and recent LLM-based fault localization techniques. Specifically, MBFL-FLIM achieves an average improvement of 44 faults in the Top-1 metric, representing a significant enhancement over baseline methods. Further evaluation confirms MBFL-FLIM's robust performance in multi-fault scenarios, with ablation experiments validating the contributions of the fine-tuning and confidence estimation components. |
| 5 | Beyond Curve Fitting: Neuro-Symbolic Agents for Context-Aware Epidemic Forecasting | 2025-11-28 | 展开Effective surveillance of hand, foot and mouth disease (HFMD) requires forecasts accounting for epidemiological patterns and contextual drivers like school calendars and weather. While classical models and recent foundation models (e.g., Chronos, TimesFM) incorporate covariates, they often lack the semantic reasoning to interpret the causal interplay between conflicting drivers. In this work, we propose a two-agent framework decoupling contextual interpretation from probabilistic forecasting. An LLM "event interpreter" processes heterogeneous signals-including school schedules, meteorological summaries, and reports-into a scalar transmission-impact signal. A neuro-symbolic core then combines this with historical case counts to produce calibrated probabilistic forecasts. We evaluate the framework on real-world HFMD datasets from Hong Kong (2023-2024) and Lishui, China (2024). Compared to traditional and foundation-model baselines, our approach achieves competitive point forecasting accuracy while providing robust 90% prediction intervals (coverage 0.85-1.00) and human-interpretable rationales. Our results suggest that structurally integrating domain knowledge through LLMs can match state-of-the-art performance while yielding context-aware forecasts that align with public health workflows. Code is available at https://github.com/jw-chae/forecast_MED . |
| 6 | GAPS: Guiding Dynamic Android Analysis with Static Path Synthesis | 2025-11-28 | 展开Dynamically resolving method reachability in Android applications remains a critical and largely unsolved problem. Despite notable advancements in GUI testing and static call graph construction, current tools are insufficient for reliably driving execution toward specific target methods, especially those not embedded in a graphical component (e.g., libraries' methods), a capability essential for tasks such as vulnerability validation, debugging, and behavioral analysis. We present GAPS (Graph-based Automated Path Synthesizer), the first system that integrates static, method-guided call graph analysis with dynamic, interaction-driven execution. GAPS performs a lightweight backward traversal of the call graph, guided by data-flow analysis, to reconstruct paths reaching the target methods. These paths are then translated into instructions that guide runtime app exploration. On the AndroTest benchmark, GAPS statically identifies paths to reach 88.24% of the target methods in just 4.27 seconds per app and dynamically reaches 57.44% of them. In contrast, state-of-the-art dynamic interaction tools show significantly lower reachability over three runs: APE, one of the best model-based GUI testers, achieves 12.82%, while GoalExplorer, a hybrid analysis tool, reaches 9.69%, and Guardian, an LLM-based UI automator, reaches 17.12%. Static analysis tools also fall short: FlowDroid and DroidReach identify paths to reach 58.81% and 9.48% of the targets, requiring 35.06 seconds and 23.46 seconds per app, respectively. Finally, an evaluation on the 50 most downloaded real-world apps demonstrates GAPS's practical utility in analyzing security-critical code under a realistic scenario. With an average static analysis time of 278.9 seconds, GAPS statically reconstructs paths to 62.03% of the target methods and dynamically reaches 59.86% of them. |
| 7 | OmniRouter: Budget and Performance Controllable Multi-LLM Routing | 2025-02-27 | 展开Large language models (LLMs) deliver superior performance but require substantial computational resources and operate with relatively low efficiency, while smaller models can efficiently handle simpler tasks with fewer resources. LLM routing is a crucial paradigm that dynamically selects the most suitable large language models from a pool of candidates to process diverse inputs, ensuring optimal resource utilization while maintaining response quality. Existing routing frameworks typically model this as a locally optimal decision-making problem, selecting the presumed best-fit LLM for each query individually, which overlooks global budget constraints, resulting in ineffective resource allocation. To tackle this problem, we introduce OmniRouter, a fundamentally controllable routing framework for multi-LLM serving. Instead of making per-query greedy choices, OmniRouter models the routing task as a constrained optimization problem, assigning models that minimize total cost while ensuring the required performance level. Specifically, a hybrid retrieval-augmented predictor is designed to predict the capabilities and costs of LLMs. After obtaining the predicted cost and performance, we utilize a constrained optimizer for cost-optimal assignments that employs Lagrangian dual decomposition with adaptive multipliers. It iteratively converges toward the globally optimal query-model allocation, dynamically balancing latency minimization against quality thresholds while adhering to heterogeneous capacity constraints. Experiments show that OmniRouter achieves up to 6.30% improvement in response accuracy while simultaneously reducing computational costs by at least 10.15% compared to competitive router baselines. The code and the dataset are available at https://github.com/dongyuanjushi/OmniRouter. |
| 序号 | 标题 | 日期 | 摘要 |
|---|---|---|---|
| 1 | Scaling HuBERT for African Languages: From Base to Large and XL | 2025-11-28 | 展开Despite recent progress in multilingual speech processing, African languages remain under-represented in both research and deployed systems, particularly when it comes to strong, open-weight encoders that transfer well under low-resource supervision. Self-supervised learning has proven especially promising in such settings, yet most publicly released models targeting African speech remain at BASE scale, leaving unanswered whether larger encoders, trained exclusively on Africa-centric audio, offer tangible benefits and how model capacity interacts with data composition. This work addresses that gap by introducing SSA-HuBERT-Large (317M parameters) and SSA-HuBERT-XL (964M parameters), the first large models trained solely on African speech, alongside a BASE size counterpart. We release these models as open weights: see https://huggingface.co/collections/Orange/african-speech-foundation-models. By conducting a carefully controlled experimental study focused exclusively on Sub-Saharan languages, covering automatic speech recognition (ASR) and language identification (LID) tasks, we demonstrate that larger architectures significantly improve performance by effectively leveraging large audio datasets. |
| 2 | HPSU: A Benchmark for Human-Level Perception in Real-World Spoken Speech Understanding | 2025-11-28 | 展开Recent advances in Speech Large Language Models (Speech LLMs) have led to great progress in speech understanding tasks such as Automatic Speech Recognition (ASR) and Speech Emotion Recognition (SER). However, whether these models can achieve human-level auditory perception, particularly in terms of their ability to comprehend latent intentions and implicit emotions in real-world spoken language, remains underexplored. To this end, we introduce the Human-level Perception in Spoken Speech Understanding (HPSU), a new benchmark for fully evaluating the human-level perceptual and understanding capabilities of Speech LLMs. HPSU comprises over 20,000 expert-validated spoken language understanding samples in English and Chinese. It establishes a comprehensive evaluation framework by encompassing a spectrum of tasks, ranging from basic speaker attribute recognition to complex inference of latent intentions and implicit emotions. To address the issues of data scarcity and high cost of manual annotation in real-world scenarios, we developed a semi-automatic annotation process. This process fuses audio, textual, and visual information to enable precise speech understanding and labeling, thus enhancing both annotation efficiency and quality. We systematically evaluate various open-source and proprietary Speech LLMs. The results demonstrate that even top-performing models still fall considerably short of human capabilities in understanding genuine spoken interactions. Consequently, HPSU will be useful for guiding the development of Speech LLMs toward human-level perception and cognition. |
| 3 | Group-Aware Partial Model Merging for Children's Automatic Speech Recognition | 2025-11-28 | 展开Automatic Speech Recognition (ASR) for children remains challenging, primarily due to large acoustic variability and limited availability of training data. While supervised fine-tuning of adult pre-trained models has shown promise, it often fails to capture group-specific characteristics variations among children. To address this, we introduce GRoup-Aware PARtial model Merging (GRAPAM), a parameter-efficient approach that combines unsupervised clustering, partial fine-tuning, and model merging. Our approach adapts adult-pre-trained models to children by first grouping the children's data based on acoustic similarity. Each group is used to partially fine-tune an adult pre-trained model, and the resulting models are merged at the parameter level. Experiments conducted on the MyST children's speech corpus indicate that GRAPAM achieves a relative improvement of 6% of Word Error Rate (WER), using the same amount of data, outperforming full fine-tuning while training fewer parameters. These results highlight the promise of model merging as a scalable and effective strategy for children's ASR. |
| 4 | Modeling Romanized Hindi and Bengali: Dataset Creation and Multilingual LLM Integration | 2025-11-27 | 展开The development of robust transliteration techniques to enhance the effectiveness of transforming Romanized scripts into native scripts is crucial for Natural Language Processing tasks, including sentiment analysis, speech recognition, information retrieval, and intelligent personal assistants. Despite significant advancements, state-of-the-art multilingual models still face challenges in handling Romanized script, where the Roman alphabet is adopted to represent the phonetic structure of diverse languages. Within the South Asian context, where the use of Romanized script for Indo-Aryan languages is widespread across social media and digital communication platforms, such usage continues to pose significant challenges for cutting-edge multilingual models. While a limited number of transliteration datasets and models are available for Indo-Aryan languages, they generally lack sufficient diversity in pronunciation and spelling variations, adequate code-mixed data for large language model (LLM) training, and low-resource adaptation. To address this research gap, we introduce a novel transliteration dataset for two popular Indo-Aryan languages, Hindi and Bengali, which are ranked as the 3rd and 7th most spoken languages worldwide. Our dataset comprises nearly 1.8 million Hindi and 1 million Bengali transliteration pairs. In addition to that, we pre-train a custom multilingual seq2seq LLM based on Marian architecture using the developed dataset. Experimental results demonstrate significant improvements compared to existing relevant models in terms of BLEU and CER metrics. |
| 5 | Do You See What I Say? Generalizable Deepfake Detection based on Visual Speech Recognition | 2025-11-27 | 展开Deepfake generation has witnessed remarkable progress, contributing to highly realistic generated images, videos, and audio. While technically intriguing, such progress has raised serious concerns related to the misuse of manipulated media. To mitigate such misuse, robust and reliable deepfake detection is urgently needed. Towards this, we propose a novel network FauxNet, which is based on pre-trained Visual Speech Recognition (VSR) features. By extracting temporal VSR features from videos, we identify and segregate real videos from manipulated ones. The holy grail in this context has to do with zero-shot detection, i.e., generalizable detection, which we focus on in this work. FauxNet consistently outperforms the state-of-the-art in this setting. In addition, FauxNet is able to attribute - distinguish between generation techniques from which the videos stem. Finally, we propose new datasets, referred to as Authentica-Vox and Authentica-HDTF, comprising about 38,000 real and fake videos in total, the latter created with six recent deepfake generation techniques. We provide extensive analysis and results on the Authentica datasets and FaceForensics++, demonstrating the superiority of FauxNet. The Authentica datasets will be made publicly available. |
| 6 | Privacy Disclosure of Similarity Rank in Speech and Language Processing | 2025-08-07 | 展开Speaker, author, and other biometric identification applications often compare a sample's similarity to a database of templates to determine the identity. Given that data may be noisy and similarity measures can be inaccurate, such a comparison may not reliably identify the true identity as the most similar. Still, even the similarity rank based on an inaccurate similarity measure can disclose private information about the true identity. We propose a methodology for quantifying the privacy disclosure of such a similarity rank by estimating its probability distribution. It is based on determining the histogram of the similarity rank of the true speaker, or when data is scarce, modeling the histogram with the beta-binomial distribution. We express the disclosure in terms of entropy (bits), such that the disclosure from independent features are additive. Our experiments demonstrate that all tested speaker and author characterizations contain personally identifying information (PII) that can aid in identification, with embeddings from speaker recognition algorithms containing the most information, followed by phone embeddings, linguistic embeddings, and fundamental frequency. Our initial experiments show that the disclosure of PII increases with the length of test samples, but it is bounded by the length of database templates. The provided metric, similarity rank disclosure, provides a way to compare the disclosure of PII between biometric features and merge them to aid identification. It can thus aid in the holistic evaluation of threats to privacy in speech and other biometric technologies. |
| 7 | Bridging Speech Emotion Recognition and Personality: Dataset and Temporal Interaction Condition Network | 2025-05-20 | 展开This study investigates the interaction between personality traits and emotion expression, exploring how personality information can improve speech emotion recognition (SER). We collect the personality annotation for the IEMOCAP dataset, making it the first speech dataset that contains both emotion and personality annotations (PA-IEMOCAP), and enabling direct integration of personality traits into SER. Statistical analysis on this dataset identified significant correlations between personality traits and emotional expressions. To extract finegrained personality features, we propose a temporal interaction condition network (TICN), in which personality features are integrated with HuBERT-based acoustic features for SER. Experiments show that incorporating ground-truth personality traits significantly enhances valence recognition, improving the concordance correlation coefficient (CCC) from 0.698 to 0.785 compared to the baseline without personality information. For practical applications in dialogue systems where personality information about the user is unavailable, we develop a front-end module of automatic personality recognition. Using these automatically predicted traits as inputs to our proposed TICN model, we achieve a CCC of 0.776 for valence recognition, representing an 11.17% relative improvement over the baseline. These findings confirm the effectiveness of personality-aware SER and provide a solid foundation for further exploration in personality-aware speech processing applications. |
| 8 | Spacewalk-18: A Benchmark for Multimodal and Long-form Procedural Video Understanding in Novel Domains | 2023-11-30 | 展开Learning from (procedural) videos has increasingly served as a pathway for embodied agents to acquire skills from human demonstrations. To do this, video understanding models must be able to obtain structured understandings, such as the temporal segmentation of a demonstration into sequences of actions and skills, and to generalize the understandings to novel environments, tasks, and problem domains. In pursuit of this goal, we introduce Spacewalk-18, a benchmark containing two tasks: (1) step recognition and (2) video question answering, over a dataset of temporally segmented and labeled tasks in International Space Station spacewalk recordings. In tandem, the two tasks quantify a model's ability to: (1) generalize to novel domains; (2) utilize long temporal context and multimodal (e.g. visual and speech) information. Our extensive experimental analysis highlights the challenges of Spacewalk-18, but also suggests best practices for domain generalization and long-form understanding. Notably, we discover a promising adaptation via summarization technique that leads to significant performance improvement without model fine-tuning. The Spacewalk-18 benchmark is released at https://brown-palm.github.io/Spacewalk-18/. |
| 序号 | 标题 | 日期 | 摘要 |
|---|---|---|---|
| 1 | Robust 3DGS-based SLAM via Adaptive Kernel Smoothing | 2025-11-28 | 展开In this paper, we challenge the conventional notion in 3DGS-SLAM that rendering quality is the primary determinant of tracking accuracy. We argue that, compared to solely pursuing a perfect scene representation, it is more critical to enhance the robustness of the rasterization process against parameter errors to ensure stable camera pose tracking. To address this challenge, we propose a novel approach that leverages a smooth kernel strategy to enhance the robustness of 3DGS-based SLAM. Unlike conventional methods that focus solely on minimizing rendering error, our core insight is to make the rasterization process more resilient to imperfections in the 3DGS parameters. We hypothesize that by allowing each Gaussian to influence a smoother, wider distribution of pixels during rendering, we can mitigate the detrimental effects of parameter noise from outlier Gaussians. This approach intentionally introduces a controlled blur to the rendered image, which acts as a regularization term, stabilizing the subsequent pose optimization. While a complete redesign of the rasterization pipeline is an ideal solution, we propose a practical and effective alternative that is readily integrated into existing 3DGS frameworks. Our method, termed Corrective Blurry KNN (CB-KNN), adaptively modifies the RGB values and locations of the K-nearest neighboring Gaussians within a local region. This dynamic adjustment generates a smoother local rendering, reducing the impact of erroneous GS parameters on the overall image. Experimental results demonstrate that our approach, while maintaining the overall quality of the scene reconstruction (mapping), significantly improves the robustness and accuracy of camera pose tracking. |
| 2 | Zero-Shot Multi-Criteria Visual Quality Inspection for Semi-Controlled Industrial Environments via Real-Time 3D Digital Twin Simulation | 2025-11-28 | 展开Early-stage visual quality inspection is vital for achieving Zero-Defect Manufacturing and minimizing production waste in modern industrial environments. However, the complexity of robust visual inspection systems and their extensive data requirements hinder widespread adoption in semi-controlled industrial settings. In this context, we propose a pose-agnostic, zero-shot quality inspection framework that compares real scenes against real-time Digital Twins (DT) in the RGB-D space. Our approach enables efficient real-time DT rendering by semantically describing industrial scenes through object detection and pose estimation of known Computer-Aided Design models. We benchmark tools for real-time, multimodal RGB-D DT creation while tracking consumption of computational resources. Additionally, we provide an extensible and hierarchical annotation strategy for multi-criteria defect detection, unifying pose labelling with logical and structural defect annotations. Based on an automotive use case featuring the quality inspection of an axial flux motor, we demonstrate the effectiveness of our framework. Our results demonstrate detection performace, achieving intersection-over-union (IoU) scores of up to 63.3% compared to ground-truth masks, even if using simple distance measurements under semi-controlled industrial conditions. Our findings lay the groundwork for future research on generalizable, low-data defect detection methods in dynamic manufacturing settings. |
| 3 | Multi-chain Graph Refinement and Selection for Reliable Reasoning in Large Language Models | 2025-11-28 | 展开The complex reasoning ability of Large Language Models (LLMs) poses a critical bottleneck for their practical applications. Test-time expansion methods such as Tree-of-Thought (ToT) and Graph-of-Thought (GoT) enhance reasoning by introducing intermediate reasoning structures, tree search, or graph-based exploration mechanisms. However, their reasoning strategies suffer from limited diversity, redundant search branches, and inadequate integration and error correction across heterogeneous reasoning paths. To address these limitations, we propose a novel reasoning framework called Multi-chain Graph Refinement & Selection (MGRS), which first generates multiple diverse reasoning trajectories for a given problem, refines candidate responses using a composite self- and cross-verification strategy, then constructs a reasoning relation graph and estimates the success rate of intermediate nodes, and finally computes cumulative success rates to select the most reliable answer and corresponding reasoning trajectory. Experimental results demonstrate that MGRS significantly advances both the reasoning capability and computational efficiency of reasoning enhancement methods. Across six benchmark datasets spanning four distinct tasks, MGRS achieves an average accuracy of 82.9%, outperforming state-of-the-art baselines by a clear margin of 2.1%. Remarkably, on the 24-point game, MGRS attains 100% accuracy for the first time, while delivering a 13.6x speed-up compared to the leading Forest of Thoughts framework. |
| 4 | Geometry-Consistent 4D Gaussian Splatting for Sparse-Input Dynamic View Synthesis | 2025-11-28 | 展开Gaussian Splatting has been considered as a novel way for view synthesis of dynamic scenes, which shows great potential in AIoT applications such as digital twins. However, recent dynamic Gaussian Splatting methods significantly degrade when only sparse input views are available, limiting their applicability in practice. The issue arises from the incoherent learning of 4D geometry as input views decrease. This paper presents GC-4DGS, a novel framework that infuses geometric consistency into 4D Gaussian Splatting (4DGS), offering real-time and high-quality dynamic scene rendering from sparse input views. While learning-based Multi-View Stereo (MVS) and monocular depth estimators (MDEs) provide geometry priors, directly integrating these with 4DGS yields suboptimal results due to the ill-posed nature of sparse-input 4D geometric optimization. To address these problems, we introduce a dynamic consistency checking strategy to reduce estimation uncertainties of MVS across spacetime. Furthermore, we propose a global-local depth regularization approach to distill spatiotemporal-consistent geometric information from monocular depths, thereby enhancing the coherent geometry and appearance learning within the 4D volume. Extensive experiments on the popular N3DV and Technicolor datasets validate the effectiveness of GC-4DGS in rendering quality without sacrificing efficiency. Notably, our method outperforms RF-DeRF, the latest dynamic radiance field tailored for sparse-input dynamic view synthesis, and the original 4DGS by 2.62dB and 1.58dB in PSNR, respectively, with seamless deployability on resource-constrained IoT edge devices. |
| 5 | DiskChunGS: Large-Scale 3D Gaussian SLAM Through Chunk-Based Memory Management | 2025-11-28 | 展开Recent advances in 3D Gaussian Splatting (3DGS) have demonstrated impressive results for novel view synthesis with real-time rendering capabilities. However, integrating 3DGS with SLAM systems faces a fundamental scalability limitation: methods are constrained by GPU memory capacity, restricting reconstruction to small-scale environments. We present DiskChunGS, a scalable 3DGS SLAM system that overcomes this bottleneck through an out-of-core approach that partitions scenes into spatial chunks and maintains only active regions in GPU memory while storing inactive areas on disk. Our architecture integrates seamlessly with existing SLAM frameworks for pose estimation and loop closure, enabling globally consistent reconstruction at scale. We validate DiskChunGS on indoor scenes (Replica, TUM-RGBD), urban driving scenarios (KITTI), and resource-constrained Nvidia Jetson platforms. Our method uniquely completes all 11 KITTI sequences without memory failures while achieving superior visual quality, demonstrating that algorithmic innovation can overcome the memory constraints that have limited previous 3DGS SLAM methods. |
| 6 | MARVO: Marine-Adaptive Radiance-aware Visual Odometry | 2025-11-28 | 展开Underwater visual localization remains challenging due to wavelength-dependent attenuation, poor texture, and non-Gaussian sensor noise. We introduce MARVO, a physics-aware, learning-integrated odometry framework that fuses underwater image formation modeling, differentiable matching, and reinforcement-learning optimization. At the front-end, we extend transformer-based feature matcher with a Physics Aware Radiance Adapter that compensates for color channel attenuation and contrast loss, yielding geometrically consistent feature correspondences under turbidity. These semi dense matches are combined with inertial and pressure measurements inside a factor-graph backend, where we formulate a keyframe-based visual-inertial-barometric estimator using GTSAM library. Each keyframe introduces (i) Pre-integrated IMU motion factors, (ii) MARVO-derived visual pose factors, and (iii) barometric depth priors, giving a full-state MAP estimate in real time. Lastly, we introduce a Reinforcement-Learningbased Pose-Graph Optimizer that refines global trajectories beyond local minima of classical least-squares solvers by learning optimal retraction actions on SE(2). |
| 7 | Improving LLM-based Ontology Matching with fine-tuning on synthetic data | 2025-11-27 | 展开Large Language Models (LLMs) are increasingly being integrated into various components of Ontology Matching pipelines. This paper investigates the capability of LLMs to perform ontology matching directly on ontology modules and generate the corresponding alignments. Furthermore, it is explored how a dedicated fine-tuning strategy can enhance the model's matching performance in a zero-shot setting. The proposed method incorporates a search space reduction technique to select relevant subsets from both source and target ontologies, which are then used to automatically construct prompts. Recognizing the scarcity of reference alignments for training, a novel LLM-based approach is introduced for generating a synthetic dataset. This process creates a corpus of ontology submodule pairs and their corresponding reference alignments, specifically designed to fine-tune an LLM for the ontology matching task. The proposed approach was evaluated on the Conference, Geolink, Enslaved, Taxon, and Hydrography datasets from the OAEI complex track. The results demonstrate that the LLM fine-tuned on the synthetically generated data exhibits superior performance compared to the non-fine-tuned base model. The key contribution is a strategy that combines automatic dataset generation with fine-tuning to effectively adapt LLMs for ontology matching tasks. |
| 8 | UAV-MM3D: A Large-Scale Synthetic Benchmark for 3D Perception of Unmanned Aerial Vehicles with Multi-Modal Data | 2025-11-27 | 展开Accurate perception of UAVs in complex low-altitude environments is critical for airspace security and related intelligent systems. Developing reliable solutions requires large-scale, accurately annotated, and multimodal data. However, real-world UAV data collection faces inherent constraints due to airspace regulations, privacy concerns, and environmental variability, while manual annotation of 3D poses and cross-modal correspondences is time-consuming and costly. To overcome these challenges, we introduce UAV-MM3D, a high-fidelity multimodal synthetic dataset for low-altitude UAV perception and motion understanding. It comprises 400K synchronized frames across diverse scenes (urban areas, suburbs, forests, coastal regions) and weather conditions (clear, cloudy, rainy, foggy), featuring multiple UAV models (micro, small, medium-sized) and five modalities - RGB, IR, LiDAR, Radar, and DVS (Dynamic Vision Sensor). Each frame provides 2D/3D bounding boxes, 6-DoF poses, and instance-level annotations, enabling core tasks related to UAVs such as 3D detection, pose estimation, target tracking, and short-term trajectory forecasting. We further propose LGFusionNet, a LiDAR-guided multimodal fusion baseline, and a dedicated UAV trajectory prediction baseline to facilitate benchmarking. With its controllable simulation environment, comprehensive scenario coverage, and rich annotations, UAV3D offers a public benchmark for advancing 3D perception of UAVs. |
| 9 | RecToM: A Benchmark for Evaluating Machine Theory of Mind in LLM-based Conversational Recommender Systems | 2025-11-27 | 展开Large Language models are revolutionizing the conversational recommender systems through their impressive capabilities in instruction comprehension, reasoning, and human interaction. A core factor underlying effective recommendation dialogue is the ability to infer and reason about users' mental states (such as desire, intention, and belief), a cognitive capacity commonly referred to as Theory of Mind. Despite growing interest in evaluating ToM in LLMs, current benchmarks predominantly rely on synthetic narratives inspired by Sally-Anne test, which emphasize physical perception and fail to capture the complexity of mental state inference in realistic conversational settings. Moreover, existing benchmarks often overlook a critical component of human ToM: behavioral prediction, the ability to use inferred mental states to guide strategic decision-making and select appropriate conversational actions for future interactions. To better align LLM-based ToM evaluation with human-like social reasoning, we propose RecToM, a novel benchmark for evaluating ToM abilities in recommendation dialogues. RecToM focuses on two complementary dimensions: Cognitive Inference and Behavioral Prediction. The former focus on understanding what has been communicated by inferring the underlying mental states. The latter emphasizes what should be done next, evaluating whether LLMs can leverage these inferred mental states to predict, select, and assess appropriate dialogue strategies. Extensive experiments on state-of-the-art LLMs demonstrate that RecToM poses a significant challenge. While the models exhibit partial competence in recognizing mental states, they struggle to maintain coherent, strategic ToM reasoning throughout dynamic recommendation dialogues, particularly in tracking evolving intentions and aligning conversational strategies with inferred mental states. |
| 10 | RefineBench: Evaluating Refinement Capability of Language Models via Checklists | 2025-11-27 | 展开Can language models (LMs) self-refine their own responses? This question is increasingly relevant as a wide range of real-world user interactions involve refinement requests. However, prior studies have largely tested LMs' refinement abilities on verifiable tasks such as competition math or symbolic reasoning with simplified scaffolds, whereas users often pose open-ended queries and provide varying degrees of feedback on what they desire. The recent advent of reasoning models that exhibit self-reflection patterns in their chains-of-thought further motivates this question. To analyze this, we introduce RefineBench, a benchmark of 1,000 challenging problems across 11 domains paired with a checklist-based evaluation framework. We evaluate two refinement modes: (1) guided refinement, where an LM is provided natural language feedback, and (2) self-refinement, where LMs attempt to improve without guidance. In the self-refinement setting, even frontier LMs such as Gemini 2.5 Pro and GPT-5 achieve modest baseline scores of 31.3% and 29.1%, respectively, and most models fail to consistently improve across iterations (e.g., Gemini-2.5-Pro gains only +1.8%, while DeepSeek-R1 declines by -0.1%). By contrast, in guided refinement, both proprietary LMs and large open-weight LMs (>70B) can leverage targeted feedback to refine responses to near-perfect levels within five turns. These findings suggest that frontier LMs require breakthroughs to self-refine their incorrect responses, and that RefineBench provides a valuable testbed for tracking progress. |
| 11 | On the Limits of Innate Planning in Large Language Models | 2025-11-26 | 展开Large language models (LLMs) achieve impressive results on many benchmarks, yet their capacity for planning and stateful reasoning remains unclear. We study these abilities directly, without code execution or other tools, using the 8-puzzle: a classic task that requires state tracking and goal-directed planning while allowing precise, step-by-step evaluation. Four models are tested under common prompting conditions (Zero-Shot, Chain-of-Thought, Algorithm-of-Thought) and with tiered corrective feedback. Feedback improves success rates for some model-prompt combinations, but many successful runs are long, computationally expensive, and indirect. We then examine the models with an external move validator that provides only valid moves. Despite this level of assistance, none of the models solve any puzzles in this setting. Qualitative analysis reveals two dominant deficits across all models: (1) brittle internal state representations, leading to frequent invalid moves, and (2) weak heuristic planning, with models entering loops or selecting actions that do not reduce the distance to the goal state. These findings indicate that, in the absence of external tools such as code interpreters, current LLMs have substantial limitations in planning and that further progress may require mechanisms for maintaining explicit state and performing structured search. |
| 12 | MortgageLLM: Domain-Adaptive Pretraining with Residual Instruction Transfer, Alignment Tuning, and Task-Specific Routing | 2025-11-26 | 展开Large Language Models (LLMs) demonstrate exceptional capabilities across general domains, yet their application to specialized sectors such as mortgage finance requires domain-specific knowledge augmentation while preserving instruction-following fidelity. We present MortgageLLM, a novel domain-specific large language model that addresses this dual challenge. It is developed using a dual-track specialization framework from a single base model (LLaMA-3.1-8B). We opted for this dual-expert approach as a single multi-task model suffers from performance trade-offs, where optimizing for structured tasks (via SFT) degrades conversational fidelity (via DPO). Our dual-track method solves this by creating two specialists, allowing each to be optimally trained for its distinct capability. Our approach applies the instruction residual technique to restore instruction-following capabilities post-domain adaptation without supervised fine-tuning. We contribute: (1) application of this residual technique to the highly specialized mortgage finance domain; (2) a dual-expert architecture combining a conversational Q&A model and a structured task model for classification and summarization; and (3) an intelligent task routing mechanism using few-shot classification performed by one of the expert models itself. We validate our approach on domain-specific benchmarks, where our final model (MLM v2) significantly outperforms the base LLaMA-3.1-8B-Instruct, achieving an LLM-as-a-Judge summarization score of 4.58 (vs. 3.99), a Q&A score of 4.09 (vs. 4.0), and a classification score of 2.6 (vs. 1.2). On semantic similarity, our model achieved a BERTScore of 0.77 for summarization (vs. 0.74), 0.68 for Q&A (vs. 0.58), and 0.75 for classification (vs. 0.73), substantially outperforming baseline approaches. |
| 13 | Image Diffusion Models Exhibit Emergent Temporal Propagation in Videos | 2025-11-25 | 展开Image diffusion models, though originally developed for image generation, implicitly capture rich semantic structures that enable various recognition and localization tasks beyond synthesis. In this work, we investigate their self-attention maps can be reinterpreted as semantic label propagation kernels, providing robust pixel-level correspondences between relevant image regions. Extending this mechanism across frames yields a temporal propagation kernel that enables zero-shot object tracking via segmentation in videos. We further demonstrate the effectiveness of test-time optimization strategies-DDIM inversion, textual inversion, and adaptive head weighting-in adapting diffusion features for robust and consistent label propagation. Building on these findings, we introduce DRIFT, a framework for object tracking in videos leveraging a pretrained image diffusion model with SAM-guided mask refinement, achieving state-of-the-art zero-shot performance on standard video object segmentation benchmarks. |
| 14 | IDSplat: Instance-Decomposed 3D Gaussian Splatting for Driving Scenes | 2025-11-24 | 展开Reconstructing dynamic driving scenes is essential for developing autonomous systems through sensor-realistic simulation. Although recent methods achieve high-fidelity reconstructions, they either rely on costly human annotations for object trajectories or use time-varying representations without explicit object-level decomposition, leading to intertwined static and dynamic elements that hinder scene separation. We present IDSplat, a self-supervised 3D Gaussian Splatting framework that reconstructs dynamic scenes with explicit instance decomposition and learnable motion trajectories, without requiring human annotations. Our key insight is to model dynamic objects as coherent instances undergoing rigid transformations, rather than unstructured time-varying primitives. For instance decomposition, we employ zero-shot, language-grounded video tracking anchored to 3D using lidar, and estimate consistent poses via feature correspondences. We introduce a coordinated-turn smoothing scheme to obtain temporally and physically consistent motion trajectories, mitigating pose misalignments and tracking failures, followed by joint optimization of object poses and Gaussian parameters. Experiments on the Waymo Open Dataset demonstrate that our method achieves competitive reconstruction quality while maintaining instance-level decomposition and generalizes across diverse sequences and view densities without retraining, making it practical for large-scale autonomous driving applications. Code will be released. |
| 15 | SatSAM2: Motion-Constrained Video Object Tracking in Satellite Imagery using Promptable SAM2 and Kalman Priors | 2025-11-23 | 展开Existing satellite video tracking methods often struggle with generalization, requiring scenario-specific training to achieve satisfactory performance, and are prone to track loss in the presence of occlusion. To address these challenges, we propose SatSAM2, a zero-shot satellite video tracker built on SAM2, designed to adapt foundation models to the remote sensing domain. SatSAM2 introduces two core modules: a Kalman Filter-based Constrained Motion Module (KFCMM) to exploit temporal motion cues and suppress drift, and a Motion-Constrained State Machine (MCSM) to regulate tracking states based on motion dynamics and reliability. To support large-scale evaluation, we propose MatrixCity Video Object Tracking (MVOT), a synthetic benchmark containing 1,500+ sequences and 157K annotated frames with diverse viewpoints, illumination, and occlusion conditions. Extensive experiments on two satellite tracking benchmarks and MVOT show that SatSAM2 outperforms both traditional and foundation model-based trackers, including SAM2 and its variants. Notably, on the OOTB dataset, SatSAM2 achieves a 5.84% AUC improvement over state-of-the-art methods. Our code and dataset will be publicly released to encourage further research. |
| 16 | RacketVision: A Multiple Racket Sports Benchmark for Unified Ball and Racket Analysis | 2025-11-21 | 展开We introduce RacketVision, a novel dataset and benchmark for advancing computer vision in sports analytics, covering table tennis, tennis, and badminton. The dataset is the first to provide large-scale, fine-grained annotations for racket pose alongside traditional ball positions, enabling research into complex human-object interactions. It is designed to tackle three interconnected tasks: fine-grained ball tracking, articulated racket pose estimation, and predictive ball trajectory forecasting. Our evaluation of established baselines reveals a critical insight for multi-modal fusion: while naively concatenating racket pose features degrades performance, a CrossAttention mechanism is essential to unlock their value, leading to trajectory prediction results that surpass strong unimodal baselines. RacketVision provides a versatile resource and a strong starting point for future research in dynamic object tracking, conditional motion forecasting, and multimodal analysis in sports. Project page at https://github.com/OrcustD/RacketVision |
| 17 | Lacking Data? No worries! How synthetic images can alleviate image scarcity in wildlife surveys: a case study with muskox (Ovibos moschatus) | 2025-11-14 | 展开Accurate population estimates are essential for wildlife management, providing critical insights into species abundance and distribution. Traditional survey methods, including visual aerial counts and GNSS telemetry tracking, are widely used to monitor muskox populations in Arctic regions. These approaches are resource intensive and constrained by logistical challenges. Advances in remote sensing, artificial intelligence, and high resolution aerial imagery offer promising alternatives for wildlife detection. Yet, the effectiveness of deep learning object detection models (ODMs) is often limited by small datasets, making it challenging to train robust ODMs for sparsely distributed species like muskoxen. This study investigates the integration of synthetic imagery (SI) to supplement limited training data and improve muskox detection in zero shot (ZS) and few-shot (FS) settings. We compared a baseline model trained on real imagery with 5 ZS and 5 FS models that incorporated progressively more SI in the training set. For the ZS models, where no real images were included in the training set, adding SI improved detection performance. As more SI were added, performance in precision, recall and F1 score increased, but eventually plateaued, suggesting diminishing returns when SI exceeded 100% of the baseline model training dataset. For FS models, combining real and SI led to better recall and slightly higher overall accuracy compared to using real images alone, though these improvements were not statistically significant. Our findings demonstrate the potential of SI to train accurate ODMs when data is scarce, offering important perspectives for wildlife monitoring by enabling rare or inaccessible species to be monitored and to increase monitoring frequency. This approach could be used to initiate ODMs without real data and refine it as real images are acquired over time. |
| 18 | Mined Prompting and Metadata-Guided Generation for Wound Care Visual Question Answering | 2025-11-13 | 展开The rapid expansion of asynchronous remote care has intensified provider workload, creating demand for AI systems that can assist clinicians in managing patient queries more efficiently. The MEDIQA-WV 2025 shared task addresses this challenge by focusing on generating free-text responses to wound care queries paired with images. In this work, we present two complementary approaches developed for the English track. The first leverages a mined prompting strategy, where training data is embedded and the top-k most similar examples are retrieved to serve as few-shot demonstrations during generation. The second approach builds on a metadata ablation study, which identified four metadata attributes that consistently enhance response quality. We train classifiers to predict these attributes for test cases and incorporate them into the generation pipeline, dynamically adjusting outputs based on prediction confidence. Experimental results demonstrate that mined prompting improves response relevance, while metadata-guided generation further refines clinical precision. Together, these methods highlight promising directions for developing AI-driven tools that can provide reliable and efficient wound care support. |
| 19 | DMA: Online RAG Alignment with Human Feedback | 2025-11-06 | 展开Retrieval-augmented generation (RAG) systems often rely on static retrieval, limiting adaptation to evolving intent and content drift. We introduce Dynamic Memory Alignment (DMA), an online learning framework that systematically incorporates multi-granularity human feedback to align ranking in interactive settings. DMA organizes document-, list-, and response-level signals into a coherent learning pipeline: supervised training for pointwise and listwise rankers, policy optimization driven by response-level preferences, and knowledge distillation into a lightweight scorer for low-latency serving. Throughout this paper, memory refers to the model's working memory, which is the entire context visible to the LLM for In-Context Learning. We adopt a dual-track evaluation protocol mirroring deployment: (i) large-scale online A/B ablations to isolate the utility of each feedback source, and (ii) few-shot offline tests on knowledge-intensive benchmarks. Online, a multi-month industrial deployment further shows substantial improvements in human engagement. Offline, DMA preserves competitive foundational retrieval while yielding notable gains on conversational QA (TriviaQA, HotpotQA). Taken together, these results position DMA as a principled approach to feedback-driven, real-time adaptation in RAG without sacrificing baseline capability. |
| 20 | BFM-Zero: A Promptable Behavioral Foundation Model for Humanoid Control Using Unsupervised Reinforcement Learning | 2025-11-06 | 展开Building Behavioral Foundation Models (BFMs) for humanoid robots has the potential to unify diverse control tasks under a single, promptable generalist policy. However, existing approaches are either exclusively deployed on simulated humanoid characters, or specialized to specific tasks such as tracking. We propose BFM-Zero, a framework that learns an effective shared latent representation that embeds motions, goals, and rewards into a common space, enabling a single policy to be prompted for multiple downstream tasks without retraining. This well-structured latent space in BFM-Zero enables versatile and robust whole-body skills on a Unitree G1 humanoid in the real world, via diverse inference methods, including zero-shot motion tracking, goal reaching, and reward optimization, and few-shot optimization-based adaptation. Unlike prior on-policy reinforcement learning (RL) frameworks, BFM-Zero builds upon recent advancements in unsupervised RL and Forward-Backward (FB) models, which offer an objective-centric, explainable, and smooth latent representation of whole-body motions. We further extend BFM-Zero with critical reward shaping, domain randomization, and history-dependent asymmetric learning to bridge the sim-to-real gap. Those key design choices are quantitatively ablated in simulation. A first-of-its-kind model, BFM-Zero establishes a step toward scalable, promptable behavioral foundation models for whole-body humanoid control. |
| 21 | Findings of the Fourth Shared Task on Multilingual Coreference Resolution: Can LLMs Dethrone Traditional Approaches? | 2025-09-22 | 展开The paper presents an overview of the fourth edition of the Shared Task on Multilingual Coreference Resolution, organized as part of the CODI-CRAC 2025 workshop. As in the previous editions, participants were challenged to develop systems that identify mentions and cluster them according to identity coreference. A key innovation of this year's task was the introduction of a dedicated Large Language Model (LLM) track, featuring a simplified plaintext format designed to be more suitable for LLMs than the original CoNLL-U representation. The task also expanded its coverage with three new datasets in two additional languages, using version 1.3 of CorefUD - a harmonized multilingual collection of 22 datasets in 17 languages. In total, nine systems participated, including four LLM-based approaches (two fine-tuned and two using few-shot adaptation). While traditional systems still kept the lead, LLMs showed clear potential, suggesting they may soon challenge established approaches in future editions. |
| 22 | Tracking the Unstable: Appearance-Guided Motion Modeling for Robust Multi-Object Tracking in UAV-Captured Videos | 2025-08-03 | 展开Multi-object tracking (MOT) aims to track multiple objects while maintaining consistent identities across frames of a given video. In unmanned aerial vehicle (UAV) recorded videos, frequent viewpoint changes and complex UAV-ground relative motion dynamics pose significant challenges, which often lead to unstable affinity measurement and ambiguous association. Existing methods typically model motion and appearance cues separately, overlooking their spatio-temporal interplay and resulting in suboptimal tracking performance. In this work, we propose AMOT, which jointly exploits appearance and motion cues through two key components: an Appearance-Motion Consistency (AMC) matrix and a Motion-aware Track Continuation (MTC) module. Specifically, the AMC matrix computes bi-directional spatial consistency under the guidance of appearance features, enabling more reliable and context-aware identity association. The MTC module complements AMC by reactivating unmatched tracks through appearance-guided predictions that align with Kalman-based predictions, thereby reducing broken trajectories caused by missed detections. Extensive experiments on three UAV benchmarks, including VisDrone2019, UAVDT, and VT-MOT-UAV, demonstrate that our AMOT outperforms current state-of-the-art methods and generalizes well in a plug-and-play and training-free manner. |
| 23 | Taming generative video models for zero-shot optical flow extraction | 2025-07-11 | 展开Extracting optical flow from videos remains a core computer vision problem. Motivated by the recent success of large general-purpose models, we ask whether frozen self-supervised video models trained only to predict future frames can be prompted, without fine-tuning, to output flow. Prior attempts to read out depth or illumination from video generators required fine-tuning; that strategy is ill-suited for flow, where labeled data is scarce and synthetic datasets suffer from a sim-to-real gap. Inspired by the Counterfactual World Model (CWM) paradigm, which can obtain point-wise correspondences by injecting a small tracer perturbation into a next-frame predictor and tracking its propagation, we extend this idea to generative video models for zero-shot flow extraction. We explore several popular architectures and find that successful zero-shot flow extraction in this manner is aided by three model properties: (1) distributional prediction of future frames (avoiding blurry or noisy outputs); (2) factorized latents that treat each spatio-temporal patch independently; and (3) random-access decoding that can condition on any subset of future pixels. These properties are uniquely present in the recently introduced Local Random Access Sequence (LRAS) architecture. Building on LRAS, we propose KL-tracing: a novel test-time inference procedure that injects a localized perturbation into the first frame, rolls out the model one step, and computes the Kullback-Leibler divergence between perturbed and unperturbed predictive distributions. Without any flow-specific fine-tuning, our method is competitive with state-of-the-art, task-specific models on the real-world TAP-Vid DAVIS benchmark and the synthetic TAP-Vid Kubric. Our results show that counterfactual prompting of controllable generative video models is an effective alternative to supervised or photometric-loss methods for high-quality flow. |
| 24 | CzechLynx: A Dataset for Individual Identification and Pose Estimation of the Eurasian Lynx | 2025-06-05 | 展开We introduce CzechLynx, the first large-scale, open-access dataset for individual identification, pose estimation, and instance segmentation of the Eurasian lynx (Lynx lynx). CzechLynx contains 39,760 camera trap images annotated with segmentation masks, identity labels, and 20-point skeletons and covers 319 unique individuals across 15 years of systematic monitoring in two geographically distinct regions: southwest Bohemia and the Western Carpathians. In addition to the real camera trap data, we provide a large complementary set of photorealistic synthetic images and a Unity-based generation pipeline with diffusion-based text-to-texture modeling, capable of producing arbitrarily large amounts of synthetic data spanning diverse environments, poses, and coat-pattern variations. To enable systematic testing across realistic ecological scenarios, we define three complementary evaluation protocols: (i) geo-aware, (ii) time-aware open-set, and (iii) time-aware closed-set, covering cross-regional and long-term monitoring settings. With the provided resources, CzechLynx offers a unique, flexible benchmark for robust evaluation of computer vision and machine learning models across realistic ecological scenarios. |
| 25 | SANSA: Unleashing the Hidden Semantics in SAM2 for Few-Shot Segmentation | 2025-05-27 | 展开Few-shot segmentation aims to segment unseen object categories from just a handful of annotated examples. This requires mechanisms that can both identify semantically related objects across images and accurately produce segmentation masks. We note that Segment Anything 2 (SAM2), with its prompt-and-propagate mechanism, offers both strong segmentation capabilities and a built-in feature matching process. However, we show that its representations are entangled with task-specific cues optimized for object tracking, which impairs its use for tasks requiring higher level semantic understanding. Our key insight is that, despite its class-agnostic pretraining, SAM2 already encodes rich semantic structure in its features. We propose SANSA (Semantically AligNed Segment Anything 2), a framework that makes this latent structure explicit, and repurposes SAM2 for few-shot segmentation through minimal task-specific modifications. SANSA achieves state-of-the-art performance on few-shot segmentation benchmarks specifically designed to assess generalization, outperforms generalist methods in the popular in-context setting, supports various prompts flexible interaction via points, boxes, or scribbles, and remains significantly faster and more compact than prior approaches. Code is available at https://github.com/ClaudiaCuttano/SANSA. |
| 26 | A Simple yet Effective Test-Time Adaptation for Zero-Shot Monocular Metric Depth Estimation | 2024-12-18 | 展开The recent development of \emph{foundation models} for monocular depth estimation such as Depth Anything paved the way to zero-shot monocular depth estimation. Since it returns an affine-invariant disparity map, the favored technique to recover the metric depth consists in fine-tuning the model. However, this stage is not straightforward, it can be costly and time-consuming because of the training and the creation of the dataset. The latter must contain images captured by the camera that will be used at test time and the corresponding ground truth. Moreover, the fine-tuning may also degrade the generalizing capacity of the original model. Instead, we propose in this paper a new method to rescale Depth Anything predictions using 3D points provided by sensors or techniques such as low-resolution LiDAR or structure-from-motion with poses given by an IMU. This approach avoids fine-tuning and preserves the generalizing power of the original depth estimation model while being robust to the noise of the sparse depth, of the camera-LiDAR calibration or of the depth model. Our experiments highlight enhancements relative to zero-shot monocular metric depth estimation methods, competitive results compared to fine-tuned approaches and a better robustness than depth completion approaches. Code available at github.com/ENSTA-U2IS-AI/depth-rescaling. |
| 27 | PoseAdapt: Sustainable Human Pose Estimation via Continual Learning Benchmarks and Toolkit | 2024-09-30 | 展开Human pose estimators are typically retrained from scratch or naively fine-tuned whenever keypoint sets, sensing modalities, or deployment domains change--an inefficient, compute-intensive practice that rarely matches field constraints. We present PoseAdapt, an open-source framework and benchmark suite for continual pose model adaptation. PoseAdapt defines domain-incremental and class-incremental tracks that simulate realistic changes in density, lighting, and sensing modality, as well as skeleton growth. The toolkit supports two workflows: (i) Strategy Benchmarking, which lets researchers implement continual learning (CL) methods as plugins and evaluate them under standardized protocols; and (ii) Model Adaptation, which allows practitioners to adapt strong pretrained models to new tasks with minimal supervision. We evaluate representative regularization-based methods in single-step and sequential settings. Benchmarks enforce a fixed lightweight backbone, no access to past data, and tight per-step budgets. This isolates adaptation strategy effects, highlighting the difficulty of maintaining accuracy under strict resource limits. PoseAdapt connects modern CL techniques with practical pose estimation needs, enabling adaptable models that improve over time without repeated full retraining. |
| 28 | Agent-OM: Leveraging LLM Agents for Ontology Matching | 2023-12-01 | 展开Ontology matching (OM) enables semantic interoperability between different ontologies and resolves their conceptual heterogeneity by aligning related entities. OM systems currently have two prevailing design paradigms: conventional knowledge-based expert systems and newer machine learning-based predictive systems. While large language models (LLMs) and LLM agents have revolutionised data engineering and have been applied creatively in many domains, their potential for OM remains underexplored. This study introduces a novel agent-powered LLM-based design paradigm for OM systems. With consideration of several specific challenges in leveraging LLM agents for OM, we propose a generic framework, namely Agent-OM (Agent for Ontology Matching), consisting of two Siamese agents for retrieval and matching, with a set of OM tools. Our framework is implemented in a proof-of-concept system. Evaluations of three Ontology Alignment Evaluation Initiative (OAEI) tracks over state-of-the-art OM systems show that our system can achieve results very close to the long-standing best performance on simple OM tasks and can significantly improve the performance on complex and few-shot OM tasks. |
| 序号 | 标题 | 日期 | 摘要 |
|---|---|---|---|
| 1 | Object-Centric Data Synthesis for Category-level Object Detection | 2025-11-28 | 展开Deep learning approaches to object detection have achieved reliable detection of specific object classes in images. However, extending a model's detection capability to new object classes requires large amounts of annotated training data, which is costly and time-consuming to acquire, especially for long-tailed classes with insufficient representation in existing datasets. Here, we introduce the object-centric data setting, when limited data is available in the form of object-centric data (multi-view images or 3D models), and systematically evaluate the performance of four different data synthesis methods to finetune object detection models on novel object categories in this setting. The approaches are based on simple image processing techniques, 3D rendering, and image diffusion models, and use object-centric data to synthesize realistic, cluttered images with varying contextual coherence and complexity. We assess how these methods enable models to achieve category-level generalization in real-world data, and demonstrate significant performance boosts within this data-constrained experimental setting. |
| 2 | Optimization and application of ultra-high field preclinical high-resolution and 3D 1H-MRSI using compressed sensing | 2025-11-28 | 展开Proton magnetic resonance spectroscopic imaging (1H-MRSI) at ultra-high field has seen an increase in usage in the preclinical field. Challenges related to long acquisition time and low concentration of brain metabolites in the rodent brain have led to the development and application of acceleration schemes for 3D-1H-MRSI, such the undersampling technique Compressed Sensing (CS). This present study aims to explore the CS tool in the context of preclinical in vivo application in order to achieve high-resolution MRSI acquisition in both 2D with an in-plane increase and 3D/multi-slice acquisition with through-plane. The parameters are explored to achieve the highest acceleration possible as a way to make 3D as time efficient as possible. Results of the parameter study showed that an acceleration factor (AF) of 4 was possible with the right sampling size of the core at the center of the k-space. With this specific set, higher matrix size resulting in sub 1 μL nominal voxel size was explored with 2D-FID-MRSI and 9 supplementary phase-encoding/slices were added to achieve 3D-FID-MRSI. The spectral quality and the metabolic maps were accurate enough in the comparison with the non-accelerated 2D-FID-MRSI, within the slice of interest. Issues related with the point spread function (PSF) were noted throughout the different usage of CS. Our work presents a robust and effective protocol to achieve 3D-1H-MRSI using CS in order to reach an acquisition time below the 30 minutes bar, with minimal technical limitations and high-quality acquisition. |
| 3 | Deep Learning for Restoring MPI System Matrices Using Simulated Training Data | 2025-11-28 | 展开Magnetic particle imaging reconstructs tracer distributions using a system matrix obtained through time-consuming, noise-prone calibration measurements. Methods for addressing imperfections in measured system matrices increasingly rely on deep neural networks, yet curated training data remain scarce. This study evaluates whether physics-based simulated system matrices can be used to train deep learning models for different system matrix restoration tasks, i.e., denoising, accelerated calibration, upsampling, and inpainting, that generalize to measured data. A large system matrices dataset was generated using an equilibrium magnetization model extended with uniaxial anisotropy. The dataset spans particle, scanner, and calibration parameters for 2D and 3D trajectories, and includes background noise injected from empty-frame measurements. For each restoration task, deep learning models were compared with classical non-learning baseline methods. The models trained solely on simulated system matrices generalized to measured data across all tasks: for denoising, DnCNN/RDN/SwinIR outperformed DCT-F baseline by >10 dB PSNR and up to 0.1 SSIM on simulations and led to perceptually better reconstuctions of real data; for 2D upsampling, SMRnet exceeded bicubic by 20 dB PSNR and 0.08 SSIM at |
| 4 | Learning to Predict Aboveground Biomass from RGB Images with 3D Synthetic Scenes | 2025-11-28 | 展开Forests play a critical role in global ecosystems by supporting biodiversity and mitigating climate change via carbon sequestration. Accurate aboveground biomass (AGB) estimation is essential for assessing carbon storage and wildfire fuel loads, yet traditional methods rely on labor-intensive field measurements or remote sensing approaches with significant limitations in dense vegetation. In this work, we propose a novel learning-based method for estimating AGB from a single ground-based RGB image. We frame this as a dense prediction task, introducing AGB density maps, where each pixel represents tree biomass normalized by the plot area and each tree's image area. We leverage the recently introduced synthetic 3D SPREAD dataset, which provides realistic forest scenes with per-image tree attributes (height, trunk and canopy diameter) and instance segmentation masks. Using these assets, we compute AGB via allometric equations and train a model to predict AGB density maps, integrating them to recover the AGB estimate for the captured scene. Our approach achieves a median AGB estimation error of 1.22 kg/m^2 on held-out SPREAD data and 1.94 kg/m^2 on a real-image dataset. To our knowledge, this is the first method to estimate aboveground biomass directly from a single RGB image, opening up the possibility for a scalable, interpretable, and cost-effective solution for forest monitoring, while also enabling broader participation through citizen science initiatives. |
| 5 | GeoWorld: Unlocking the Potential of Geometry Models to Facilitate High-Fidelity 3D Scene Generation | 2025-11-28 | 展开Previous works leveraging video models for image-to-3D scene generation tend to suffer from geometric distortions and blurry content. In this paper, we renovate the pipeline of image-to-3D scene generation by unlocking the potential of geometry models and present our GeoWorld. Instead of exploiting geometric information obtained from a single-frame input, we propose to first generate consecutive video frames and then take advantage of the geometry model to provide full-frame geometry features, which contain richer information than single-frame depth maps or camera embeddings used in previous methods, and use these geometry features as geometrical conditions to aid the video generation model. To enhance the consistency of geometric structures, we further propose a geometry alignment loss to provide the model with real-world geometric constraints and a geometry adaptation module to ensure the effective utilization of geometry features. Extensive experiments show that our GeoWorld can generate high-fidelity 3D scenes from a single image and a given camera trajectory, outperforming prior methods both qualitatively and quantitatively. Project Page: https://peaes.github.io/GeoWorld/. |
| 6 | Electronic Structure and Dynamical Correlations in Antiferromagnetic BiFeO$_3$ | 2025-11-28 | 展开We study the electronic structure and dynamical correlations in antiferromagnetic BiFeO$_3$, a prototypical room-temperature multiferroic, using a variety of static and dynamical first-principles methods. Conventional static Hubbard corrections (DFT+$U$, DFT+$U$+$V$) incorrectly predict a deep-valence Fe |
| 7 | Fast Multi-view Consistent 3D Editing with Video Priors | 2025-11-28 | 展开Text-driven 3D editing enables user-friendly 3D object or scene editing with text instructions. Due to the lack of multi-view consistency priors, existing methods typically resort to employing 2D generation or editing models to process each view individually, followed by iterative 2D-3D-2D updating. However, these methods are not only time-consuming but also prone to over-smoothed results because the different editing signals gathered from different views are averaged during the iterative process. In this paper, we propose generative Video Prior based 3D Editing (ViP3DE) to employ the temporal consistency priors from pre-trained video generation models for multi-view consistent 3D editing in a single forward pass. Our key insight is to condition the video generation model on a single edited view to generate other consistent edited views for 3D updating directly, thereby bypassing the iterative editing paradigm. Since 3D updating requires edited views to be paired with specific camera poses, we propose motion-preserved noise blending for the video model to generate edited views at predefined camera poses. In addition, we introduce geometry-aware denoising to further enhance multi-view consistency by integrating 3D geometric priors into video models. Extensive experiments demonstrate that our proposed ViP3DE can achieve high-quality 3D editing results even within a single forward pass, significantly outperforming existing methods in both editing quality and speed. |
| 8 | Geodiffussr: Generative Terrain Texturing with Elevation Fidelity | 2025-11-28 | 展开Large-scale terrain generation remains a labor-intensive task in computer graphics. We introduce Geodiffussr, a flow-matching pipeline that synthesizes text-guided texture maps while strictly adhering to a supplied Digital Elevation Map (DEM). The core mechanism is multi-scale content aggregation (MCA): DEM features from a pretrained encoder are injected into UNet blocks at multiple resolutions to enforce global-to-local elevation consistency. Compared with a non-MCA baseline, MCA markedly improves visual fidelity and strengthens height-appearance coupling (FID |
| 9 | HMR3D: Hierarchical Multimodal Representation for 3D Scene Understanding with Large Vision-Language Model | 2025-11-28 | 展开Recent advances in large vision-language models (VLMs) have shown significant promise for 3D scene understanding. Existing VLM-based approaches typically align 3D scene features with the VLM's embedding space. However, this implicit alignment often yields suboptimal performance due to the scarcity of 3D data and the inherent complexity of spatial relationships in 3D environments. To address these limitations, we propose a novel hierarchical multimodal representation for 3D scene reasoning that explicitly aligns with VLMs at the input space by leveraging both multi-view images and text descriptions. The text descriptions capture spatial relationships by referencing the 3D coordinates of detected objects, while the multi-view images include a top-down perspective and four directional views (forward, left, right, and backward), ensuring comprehensive scene coverage. Additionally, we introduce a hierarchical feature representation that aggregates patch-level image features into view-level and scene-level representations, enabling the model to reason over both local and global scene context. Experimental results on both situated 3D Q&A and general 3D Q&A benchmarks demonstrate the effectiveness of our approach. |
| 10 | ITS3D: Inference-Time Scaling for Text-Guided 3D Diffusion Models | 2025-11-27 | 展开We explore inference-time scaling in text-guided 3D diffusion models to enhance generative quality without additional training. To this end, we introduce ITS3D, a framework that formulates the task as an optimization problem to identify the most effective Gaussian noise input. The framework is driven by a verifier-guided search algorithm, where the search algorithm iteratively refines noise candidates based on verifier feedback. To address the inherent challenges of 3D generation, we introduce three techniques for improved stability, efficiency, and exploration capability. 1) Gaussian normalization is applied to stabilize the search process. It corrects distribution shifts when noise candidates deviate from a standard Gaussian distribution during iterative updates. 2) The high-dimensional nature of the 3D search space increases computational complexity. To mitigate this, a singular value decomposition-based compression technique is employed to reduce dimensionality while preserving effective search directions. 3) To further prevent convergence to suboptimal local minima, a singular space reset mechanism dynamically updates the search space based on diversity measures. Extensive experiments demonstrate that ITS3D enhances text-to-3D generation quality, which shows the potential of computationally efficient search methods in generative processes. The source code is available at https://github.com/ZhenglinZhou/ITS3D. |
| 11 | Wukong's 72 Transformations: High-fidelity Textured 3D Morphing via Flow Models | 2025-11-27 | 展开We present WUKONG, a novel training-free framework for high-fidelity textured 3D morphing that takes a pair of source and target prompts (image or text) as input. Unlike conventional methods -- which rely on manual correspondence matching and deformation trajectory estimation (limiting generalization and requiring costly preprocessing) -- WUKONG leverages the generative prior of flow-based transformers to produce high-fidelity 3D transitions with rich texture details. To ensure smooth shape transitions, we exploit the inherent continuity of flow-based generative processes and formulate morphing as an optimal transport barycenter problem. We further introduce a sequential initialization strategy to prevent abrupt geometric distortions and preserve identity coherence. For faithful texture preservation, we propose a similarity-guided semantic consistency mechanism that selectively retains high-frequency details and enables precise control over blending dynamics. This avoids common artifacts like oversmoothing while maintaining semantic fidelity. Extensive quantitative and qualitative evaluations demonstrate that WUKONG significantly outperforms state-of-the-art methods, achieving superior results across diverse geometry and texture variations. |
| 12 | PixelDiT: Pixel Diffusion Transformers for Image Generation | 2025-11-25 | 展开Latent-space modeling has been the standard for Diffusion Transformers (DiTs). However, it relies on a two-stage pipeline where the pretrained autoencoder introduces lossy reconstruction, leading to error accumulation while hindering joint optimization. To address these issues, we propose PixelDiT, a single-stage, end-to-end model that eliminates the need for the autoencoder and learns the diffusion process directly in the pixel space. PixelDiT adopts a fully transformer-based architecture shaped by a dual-level design: a patch-level DiT that captures global semantics and a pixel-level DiT that refines texture details, enabling efficient training of a pixel-space diffusion model while preserving fine details. Our analysis reveals that effective pixel-level token modeling is essential to the success of pixel diffusion. PixelDiT achieves 1.61 FID on ImageNet 256x256, surpassing existing pixel generative models by a large margin. We further extend PixelDiT to text-to-image generation and pretrain it at the 1024x1024 resolution in pixel space. It achieves 0.74 on GenEval and 83.5 on DPG-bench, approaching the best latent diffusion models. |
| 13 | Material-informed Gaussian Splatting for 3D World Reconstruction in a Digital Twin | 2025-11-25 | 展开3D reconstruction for Digital Twins often relies on LiDAR-based methods, which provide accurate geometry but lack the semantics and textures naturally captured by cameras. Traditional LiDAR-camera fusion approaches require complex calibration and still struggle with certain materials like glass, which are visible in images but poorly represented in point clouds. We propose a camera-only pipeline that reconstructs scenes using 3D Gaussian Splatting from multi-view images, extracts semantic material masks via vision models, converts Gaussian representations to mesh surfaces with projected material labels, and assigns physics-based material properties for accurate sensor simulation in modern graphics engines and simulators. This approach combines photorealistic reconstruction with physics-based material assignment, providing sensor simulation fidelity comparable to LiDAR-camera fusion while eliminating hardware complexity and calibration requirements. We validate our camera-only method using an internal dataset from an instrumented test vehicle, leveraging LiDAR as ground truth for reflectivity validation alongside image similarity metrics. |
| 14 | Text-guided Controllable Diffusion for Realistic Camouflage Images Generation | 2025-11-25 | 展开Camouflage Images Generation (CIG) is an emerging research area that focuses on synthesizing images in which objects are harmoniously blended and exhibit high visual consistency with their surroundings. Existing methods perform CIG by either fusing objects into specific backgrounds or outpainting the surroundings via foreground object-guided diffusion. However, they often fail to obtain natural results because they overlook the logical relationship between camouflaged objects and background environments. To address this issue, we propose CT-CIG, a Controllable Text-guided Camouflage Images Generation method that produces realistic and logically plausible camouflage images. Leveraging Large Visual Language Models (VLM), we design a Camouflage-Revealing Dialogue Mechanism (CRDM) to annotate existing camouflage datasets with high-quality text prompts. Subsequently, the constructed image-prompt pairs are utilized to finetune Stable Diffusion, incorporating a lightweight controller to guide the location and shape of camouflaged objects for enhanced camouflage scene fitness. Moreover, we design a Frequency Interaction Refinement Module (FIRM) to capture high-frequency texture features, facilitating the learning of complex camouflage patterns. Extensive experiments, including CLIPScore evaluation and camouflage effectiveness assessment, demonstrate the semantic alignment of our generated text prompts and CT-CIG's ability to produce photorealistic camouflage images. |
| 15 | Yo'City: Personalized and Boundless 3D Realistic City Scene Generation via Self-Critic Expansion | 2025-11-24 | 展开Realistic 3D city generation is fundamental to a wide range of applications, including virtual reality and digital twins. However, most existing methods rely on training a single diffusion model, which limits their ability to generate personalized and boundless city-scale scenes. In this paper, we present Yo'City, a novel agentic framework that enables user-customized and infinitely expandable 3D city generation by leveraging the reasoning and compositional capabilities of off-the-shelf large models. Specifically, Yo'City first conceptualize the city through a top-down planning strategy that defines a hierarchical "City-District-Grid" structure. The Global Planner determines the overall layout and potential functional districts, while the Local Designer further refines each district with detailed grid-level descriptions. Subsequently, the grid-level 3D generation is achieved through a "produce-refine-evaluate" isometric image synthesis loop, followed by image-to-3D generation. To simulate continuous city evolution, Yo'City further introduces a user-interactive, relationship-guided expansion mechanism, which performs scene graph-based distance- and semantics-aware layout optimization, ensuring spatially coherent city growth. To comprehensively evaluate our method, we construct a diverse benchmark dataset and design six multi-dimensional metrics that assess generation quality from the perspectives of semantics, geometry, texture, and layout. Extensive experiments demonstrate that Yo'City consistently outperforms existing state-of-the-art methods across all evaluation aspects. |
| 16 | Refracting Reality: Generating Images with Realistic Transparent Objects | 2025-11-21 | 展开Generative image models can produce convincingly real images, with plausible shapes, textures, layouts and lighting. However, one domain in which they perform notably poorly is in the synthesis of transparent objects, which exhibit refraction, reflection, absorption and scattering. Refraction is a particular challenge, because refracted pixel rays often intersect with surfaces observed in other parts of the image, providing a constraint on the color. It is clear from inspection that generative models have not distilled the laws of optics sufficiently well to accurately render refractive objects. In this work, we consider the problem of generating images with accurate refraction, given a text prompt. We synchronize the pixels within the object's boundary with those outside by warping and merging the pixels using Snell's Law of Refraction, at each step of the generation trajectory. For those surfaces that are not directly observed in the image, but are visible via refraction or reflection, we recover their appearance by synchronizing the image with a second generated image -- a panorama centered at the object -- using the same warping and merging procedure. We demonstrate that our approach generates much more optically-plausible images that respect the physical constraints. |
| 17 | DensiCrafter: Physically-Constrained Generation and Fabrication of Self-Supporting Hollow Structures | 2025-11-12 | 展开The rise of 3D generative models has enabled automatic 3D geometry and texture synthesis from multimodal inputs (e.g., text or images). However, these methods often ignore physical constraints and manufacturability considerations. In this work, we address the challenge of producing 3D designs that are both lightweight and self-supporting. We present DensiCrafter, a framework for generating lightweight, self-supporting 3D hollow structures by optimizing the density field. Starting from coarse voxel grids produced by Trellis, we interpret these as continuous density fields to optimize and introduce three differentiable, physically constrained, and simulation-free loss terms. Additionally, a mass regularization penalizes unnecessary material, while a restricted optimization domain preserves the outer surface. Our method seamlessly integrates with pretrained Trellis-based models (e.g., Trellis, DSO) without any architectural changes. In extensive evaluations, we achieve up to 43% reduction in material mass on the text-to-3D task. Compared to state-of-the-art baselines, our method could improve the stability and maintain high geometric fidelity. Real-world 3D-printing experiments confirm that our hollow designs can be reliably fabricated and could be self-supporting. |
| 18 | STAR-Bench: Probing Deep Spatio-Temporal Reasoning as Audio 4D Intelligence | 2025-10-28 | 展开Despite rapid progress in Multi-modal Large Language Models and Large Audio-Language Models, existing audio benchmarks largely test semantics that can be recovered from text captions, masking deficits in fine-grained perceptual reasoning. We formalize audio 4D intelligence that is defined as reasoning over sound dynamics in time and 3D space, and introduce STAR-Bench to measure it. STAR-Bench combines a Foundational Acoustic Perception setting (six attributes under absolute and relative regimes) with a Holistic Spatio-Temporal Reasoning setting that includes segment reordering for continuous and discrete processes and spatial tasks spanning static localization, multi-source relations, and dynamic trajectories. Our data curation pipeline uses two methods to ensure high-quality samples. For foundational tasks, we use procedurally synthesized and physics-simulated audio. For holistic data, we follow a four-stage process that includes human annotation and final selection based on human performance. Unlike prior benchmarks where caption-only answering reduces accuracy slightly, STAR-Bench induces far larger drops (-31.5% temporal, -35.2% spatial), evidencing its focus on linguistically hard-to-describe cues. Evaluating 19 models reveals substantial gaps compared with humans and a capability hierarchy: closed-source models are bottlenecked by fine-grained perception, while open-source models lag across perception, knowledge, and reasoning. Our STAR-Bench provides critical insights and a clear path forward for developing future models with a more robust understanding of the physical world. |
| 19 | ControlEvents: Controllable Synthesis of Event Camera Datawith Foundational Prior from Image Diffusion Models | 2025-09-26 | 展开In recent years, event cameras have gained significant attention due to their bio-inspired properties, such as high temporal resolution and high dynamic range. However, obtaining large-scale labeled ground-truth data for event-based vision tasks remains challenging and costly. In this paper, we present ControlEvents, a diffusion-based generative model designed to synthesize high-quality event data guided by diverse control signals such as class text labels, 2D skeletons, and 3D body poses. Our key insight is to leverage the diffusion prior from foundation models, such as Stable Diffusion, enabling high-quality event data generation with minimal fine-tuning and limited labeled data. Our method streamlines the data generation process and significantly reduces the cost of producing labeled event datasets. We demonstrate the effectiveness of our approach by synthesizing event data for visual recognition, 2D skeleton estimation, and 3D body pose estimation. Our experiments show that the synthesized labeled event data enhances model performance in all tasks. Additionally, our approach can generate events based on unseen text labels during training, illustrating the powerful text-based generation capabilities inherited from foundation models. |
| 20 | OpenDance: Multimodal Controllable 3D Dance Generation with Large-scale Internet Data | 2025-06-09 | 展开Music-driven 3D dance generation offers significant creative potential, yet practical applications demand versatile and multimodal control. As the highly dynamic and complex human motion covering various styles and genres, dance generation requires satisfying diverse conditions beyond just music (e.g., spatial trajectories, keyframe gestures, or style descriptions). However, the absence of a large-scale and richly annotated dataset severely hinders progress. In this paper, we build OpenDanceSet, an extensive human dance dataset comprising over 100 hours across 14 genres and 147 subjects. Each sample has rich annotations to facilitate robust cross-modal learning: 3D motion, paired music, 2D keypoints, trajectories, and expert-annotated text descriptions. Furthermore, we propose OpenDanceNet, a unified masked modeling framework for controllable dance generation, including a disentangled auto-encoder and a multimodal joint-prediction Transformer. OpenDanceNet supports generation conditioned on music and arbitrary combinations of text, keypoints, or trajectories. Comprehensive experiments demonstrate that our work achieves high-fidelity synthesis with strong diversity and realistic physical contacts, while also offering flexible control over spatial and stylistic conditions. Project Page: https://open-dance.github.io |
| 21 | CzechLynx: A Dataset for Individual Identification and Pose Estimation of the Eurasian Lynx | 2025-06-05 | 展开We introduce CzechLynx, the first large-scale, open-access dataset for individual identification, pose estimation, and instance segmentation of the Eurasian lynx (Lynx lynx). CzechLynx contains 39,760 camera trap images annotated with segmentation masks, identity labels, and 20-point skeletons and covers 319 unique individuals across 15 years of systematic monitoring in two geographically distinct regions: southwest Bohemia and the Western Carpathians. In addition to the real camera trap data, we provide a large complementary set of photorealistic synthetic images and a Unity-based generation pipeline with diffusion-based text-to-texture modeling, capable of producing arbitrarily large amounts of synthetic data spanning diverse environments, poses, and coat-pattern variations. To enable systematic testing across realistic ecological scenarios, we define three complementary evaluation protocols: (i) geo-aware, (ii) time-aware open-set, and (iii) time-aware closed-set, covering cross-regional and long-term monitoring settings. With the provided resources, CzechLynx offers a unique, flexible benchmark for robust evaluation of computer vision and machine learning models across realistic ecological scenarios. |
| 22 | IntrinsiX: High-Quality PBR Generation using Image Priors | 2025-04-01 | 展开We introduce IntrinsiX, a novel method that generates high-quality intrinsic images from text description. In contrast to existing text-to-image models whose outputs contain baked-in scene lighting, our approach predicts physically-based rendering (PBR) maps. This enables the generated outputs to be used for content creation scenarios in core graphics applications that facilitate re-lighting, editing, and texture generation tasks. In order to train our generator, we exploit strong image priors, and pre-train separate models for each PBR material component (albedo, roughness, metallic, normals). We then align these models with a new cross-intrinsic attention formulation that concatenates key and value features in a consistent fashion. This allows us to exchange information between each output modality and to obtain semantically coherent PBR predictions. To ground each intrinsic component, we propose a rendering loss which provides image-space signals to constrain the model, thus facilitating sharp details also in the output BRDF properties. Our results demonstrate detailed intrinsic generation with strong generalization capabilities that outperforms existing intrinsic image decomposition methods used with generated images by a significant margin. Finally, we show a series of applications, including re-lighting, editing, and text-conditioned room-scale PBR texture generation. |
| 23 | A Simple yet Effective Test-Time Adaptation for Zero-Shot Monocular Metric Depth Estimation | 2024-12-18 | 展开The recent development of \emph{foundation models} for monocular depth estimation such as Depth Anything paved the way to zero-shot monocular depth estimation. Since it returns an affine-invariant disparity map, the favored technique to recover the metric depth consists in fine-tuning the model. However, this stage is not straightforward, it can be costly and time-consuming because of the training and the creation of the dataset. The latter must contain images captured by the camera that will be used at test time and the corresponding ground truth. Moreover, the fine-tuning may also degrade the generalizing capacity of the original model. Instead, we propose in this paper a new method to rescale Depth Anything predictions using 3D points provided by sensors or techniques such as low-resolution LiDAR or structure-from-motion with poses given by an IMU. This approach avoids fine-tuning and preserves the generalizing power of the original depth estimation model while being robust to the noise of the sparse depth, of the camera-LiDAR calibration or of the depth model. Our experiments highlight enhancements relative to zero-shot monocular metric depth estimation methods, competitive results compared to fine-tuned approaches and a better robustness than depth completion approaches. Code available at github.com/ENSTA-U2IS-AI/depth-rescaling. |
| 序号 | 标题 | 日期 | 摘要 |
|---|---|---|---|
| 1 | All for One and One for All: Program Logics for Exploiting Internal Determinism in Parallel Programs | 2025-11-28 | 展开Nondeterminism makes parallel programs challenging to write and reason about. To avoid these challenges, researchers have developed techniques for internally deterministic parallel programming, in which the steps of a parallel computation proceed in a deterministic way. Internal determinism is useful because it lets a programmer reason about a program as if it executed in a sequential order. However, no verification framework exists to exploit this property and simplify formal reasoning about internally deterministic programs. To capture the essence of why internally deterministic programs should be easier to reason about, this paper defines a property called schedule-independent safety. A program satisfies schedule-independent safety, if, to show that the program is safe across all orderings, it suffices to show that one terminating execution of the program is safe. We then present a separation logic called Musketeer for proving that a program satisfies schedule-independent safety. Once a parallel program has been shown to satisfy schedule-independent safety, we can verify it with a new logic called Angelic, which allows one to dynamically select and verify just one sequential ordering of the program. Using Musketeer, we prove the soundness of MiniDet, an affine type system for enforcing internal determinism. MiniDet supports several core algorithmic primitives for internally deterministic programming that have been identified in the research literature, including a deterministic version of a concurrent hash set. Because any syntactically well-typed MiniDet program satisfies schedule-independent safety, we can apply Angelic to verify such programs. All results in this paper have been verified in Rocq using the Iris separation logic framework. |
| 2 | AI for software engineering: from probable to provable | 2025-11-28 | 展开Vibe coding, the much-touted use of AI techniques for programming, faces two overwhelming obstacles: the difficulty of specifying goals ("prompt engineering" is a form of requirements engineering, one of the toughest disciplines of software engineering); and the hallucination phenomenon. Programs are only useful if they are correct or very close to correct. The solution? Combine the creativity of artificial intelligence with the rigor of formal specification methods and the power of formal program verification, supported by modern proof tools. |
| 3 | Automated Generation of MDPs Using Logic Programming and LLMs for Robotic Applications | 2025-11-28 | 展开We present a novel framework that integrates Large Language Models (LLMs) with automated planning and formal verification to streamline the creation and use of Markov Decision Processes (MDP). Our system leverages LLMs to extract structured knowledge in the form of a Prolog knowledge base from natural language (NL) descriptions. It then automatically constructs an MDP through reachability analysis, and synthesises optimal policies using the Storm model checker. The resulting policy is exported as a state-action table for execution. We validate the framework in three human-robot interaction scenarios, demonstrating its ability to produce executable policies with minimal manual effort. This work highlights the potential of combining language models with formal methods to enable more accessible and scalable probabilistic planning in robotics. |
| 4 | Conformal blocks of Wess-Zumino-Witten model from its free-field representation | 2025-11-28 | 展开A powerful approach to the celebrated Wess-Zumino-Witten (WZW) model is provided by its free-field realization. However, explicit calculations of conformal blocks are not described in the literature in full detail. We begin this study with the simplest cases of the |
| 5 | No-go theorems for sequential preparation of two-dimensional chiral states via channel-state correspondence | 2025-11-24 | 展开We investigate whether sequential unitary circuits can prepare two-dimensional chiral states, using a correspondence between sequentially prepared states, isometric tensor network states, and one-dimensional quantum channel circuits. We establish two no-go theorems, one for Gaussian fermion systems and one for generic interacting systems. In Gaussian fermion systems, the correspondence relates the defining features of chiral wave functions in their entanglement spectrum to the algebraic decaying correlations in the steady state of channel dynamics. We establish the no-go theorem by proving that local channel dynamics with translational invariance cannot support such correlations. As a direct implication, two-dimensional Gaussian fermion isometric tensor network states cannot support algebraically decaying correlations in all directions or represent a chiral state. In generic interacting systems, we establish a no-go theorem by showing that the state prepared by sequential circuits cannot host the tripartite entanglement of a chiral state due to the constraints from causality. |
| 6 | Geometric Rough Paths above Mixed Fractional Brownian Motion | 2025-11-24 | 展开This paper establishes a comprehensive theory of geometric rough paths for mixed fractional Brownian motion (MFBM) and its generalized multi-component extensions. We prove that for a generalized MFBM of the form |
| 7 | Phase transitions in two-component Bose-Einstein condensates with Rabi frequency (II): The De Giorgi conjecture for the nonlocal problem in $\mathbb{R}^{2}$ or $\mathbb{R}^{3}$ | 2025-11-21 | 展开In this series of papers, we investigate coupled systems arising in the study of two-component Bose-Einstein condensates, and we establish classification results for solutions of De Giorgi conjecture type. In the present (second) paper of the series, we focus on the nonlocal problem of the form \begin{equation*} \left{\begin{aligned} (-Δ)^{s}u+u(u^{2}+v^{2}-1)+v(αuv-ω)=0, (-Δ)^{s}v+v(u^{2}+v^{2}-1)+u(αuv-ω)=0, \end{aligned} \right. \end{equation*} which models the stationary states of Rabi-coupled condensates with inter- and intra-species interactions. We prove that for |
| 8 | Graph Rewriting Language as a Platform for Quantum Diagrammatic Calculi | 2025-11-19 | 展开Systematic discovery of optimization paths in quantum circuit simplification remains a challenge. Today, ZX-calculus, a computing model for quantum circuit transformation, is attracting attention for its highly abstract graph-based approach. Whereas existing tools such as PyZX and Quantomatic offer domain-specific support for quantum circuit optimization, visualization and theorem-proving, we present a complementary approach using LMNtal, a general-purpose hierarchical graph rewriting language, to establish a diagrammatic transformation and verification platform with model checking. Our methodology shows three advantages: (1) manipulation of ZX-diagrams through native graph transformation rules, enabling direct implementation of basic rules; (2) quantified pattern matching via QLMNtal extensions, greatly simplifying rule specification; and (3) interactive visualization and validation of optimization paths through state space exploration. Through case studies, we demonstrate how our framework helps understand optimization paths and design new algorithms and strategies. This suggests that the declarative language LMNtal and its toolchain could serve as a new platform to investigate quantum circuit transformation from a different perspective. |
| 9 | An Operational Kardashev-Style Scale for Autonomous AI - Towards AGI and Superintelligence | 2025-11-17 | 展开We propose a Kardashev-inspired yet operational Autonomous AI (AAI) Scale that measures the progression from fixed robotic process automation (AAI-0) to full artificial general intelligence (AAI-4) and beyond. Unlike narrative ladders, our scale is multi-axis and testable. We define ten capability axes (Autonomy, Generality, Planning, Memory/Persistence, Tool Economy, Self-Revision, Sociality/Coordination, Embodiment, World-Model Fidelity, Economic Throughput) aggregated by a composite AAI-Index (a weighted geometric mean). We introduce a measurable Self-Improvement Coefficient |
| 10 | The free Banach $f$-algebra generated by a Banach space | 2025-11-17 | 展开We construct and analyze the free Banach |
| 11 | Spark-Prover-X1: Formal Theorem Proving Through Diverse Data Training | 2025-11-17 | 展开Large Language Models (LLMs) have shown significant promise in automated theorem proving, yet progress is often constrained by the scarcity of diverse and high-quality formal language data. To address this issue, we introduce Spark-Prover-X1, a 7B parameter model trained via an three-stage framework designed to unlock the reasoning potential of more accessible and moderately-sized LLMs. The first stage infuses deep knowledge through continuous pre-training on a broad mathematical corpus, enhanced by a suite of novel data tasks. Key innovation is a "CoT-augmented state prediction" task to achieve fine-grained reasoning. The second stage employs Supervised Fine-tuning (SFT) within an expert iteration loop to specialize both the Spark-Prover-X1-7B and Spark-Formalizer-X1-7B models. Finally, a targeted round of Group Relative Policy Optimization (GRPO) is applied to sharpen the prover's capabilities on the most challenging problems. To facilitate robust evaluation, particularly on problems from real-world examinations, we also introduce ExamFormal-Bench, a new benchmark dataset of 402 formal problems. Experimental results demonstrate that Spark-Prover achieves state-of-the-art performance among similarly-sized open-source models within the "Whole-Proof Generation" paradigm. It shows exceptional performance on difficult competition benchmarks, notably solving 27 problems on PutnamBench (pass@32) and achieving 24.0% on CombiBench (pass@32). Our work validates that this diverse training data and progressively refined training pipeline provides an effective path for enhancing the formal reasoning capabilities of lightweight LLMs. Both Spark-Prover-X1-7B and Spark-Formalizer-X1-7B, along with the ExamFormal-Bench dataset, are made publicly available at: https://www.modelscope.cn/organization/iflytek, https://gitcode.com/ifly_opensource. |
| 12 | An Automated Theorem Generator with Theoretical Foundation Based on Rectangular Standard Contradiction | 2025-11-06 | 展开Currently, there is a lack of rigorous theoretical system for systematically generating non-trivial and logically valid theorems. Addressing this critical gap, this paper conducts research to propose a novel automated theorem generation theory and tool. Based on the concept of standard contradiction which possesses unique deductive advantages, this paper defines and proves, for the first time, a new logical structure known as rectangular standard contradiction. Centered on this structure, a complete Automated Theorem Generation (ATG) theory is put forward. Theoretical proofs clarify two core properties of rectangular standard contradiction: first, it is a standard contradiction (necessarily unsatisfiable); second, it exhibits non-redundancy (the remaining clause set becomes satisfiable after removing any clause). Leveraging these properties, this paper proves that partitioning a rectangular standard contradiction into a premise subset |
| 13 | Learning Interestingness in Automated Mathematical Theory Formation | 2025-11-05 | 展开We take two key steps in automating the open-ended discovery of new mathematical theories, a grand challenge in artificial intelligence. First, we introduce |
| 14 | A Compositional Account of Generalized Reversible Computing | 2025-11-04 | 展开We develop a compositional framework for generalized reversible computing using copy-discard categories and resource theories. We introduce partitioned matrices between partitioned sets as subdistribution matrices which preserve the equivalence relation of its domain. We model computational and physical transformations as subdistribution matrices over the category of sets and partitioned matrices on partitioned sets, respectively. We show that the interactions between the physical and computational transformations are governed by an aggregation functor whose functoriality and monoidality we deduce from general principles of the formal theory of monads. We study the associated copy-discard structures, in particular, general conditions for determinism and partial invertibility. We then define several notions of entropies that we use to state and prove the fundamental theorem of generalized reversible computing. |
| 15 | RLMEval: Evaluating Research-Level Neural Theorem Proving | 2025-10-29 | 展开Despite impressive results on curated benchmarks, the practical impact of large language models (LLMs) on research-level neural theorem proving and proof autoformalization is still limited. We introduce RLMEval, an evaluation suite for these tasks, focusing on research-level mathematics from real-world Lean formalization projects. RLMEval targets the evaluation of neural theorem proving and proof autoformalization on challenging research-level theorems by leveraging real Lean Blueprint formalization projects. Our evaluation of state-of-the-art models on RLMEval, comprising 613 theorems from 6 Lean projects, reveals a significant gap: progress on existing benchmarks does not readily translate to these more realistic settings, with the best model achieving only a 10.3 % pass rate. RLMEval provides a new, challenging benchmark designed to guide and accelerate progress in automated reasoning for formal mathematics. |
| 16 | A Logic for Approximate Quantitative Reasoning about Quantum Circuits | 2025-07-18 | 展开Reasoning about quantum programs remains a fundamental challenge, regardless of the programming model or computational paradigm. Despite extensive research, existing verification techniques are insufficient -- even for quantum circuits, a deliberately restricted model that lacks classical control, but still underpins many current quantum algorithms. Many existing formal methods require exponential time and space to represent and manipulate (representations of) assertions and judgments, making them impractical for quantum circuits with many qubits. This paper presents a logic for reasoning in such settings, called SAQR-QC. The logic supports {S}calable but {A}pproximate {Q}uantitative {R}easoning about {Q}uantum {C}ircuits, whence the name. SAQR-QC has three characteristics: (i) some (deliberate) loss of precision is built into it; (ii) it has a mechanism to help the accumulated loss of precision during a sequence of reasoning steps remain small; and (iii) most importantly, to make reasoning scalable, every reasoning step is local -- i.e., it involves just a small number of qubits. We demonstrate the effectiveness of SAQR-QC via two case studies: the verification of GHZ circuits involving non-Clifford gates, and the analysis of quantum phase estimation -- a core subroutine in Shor's factoring algorithm. |
| 17 | Prover Agent: An Agent-Based Framework for Formal Mathematical Proofs | 2025-06-24 | 展开We present Prover Agent, a novel AI agent for automated theorem proving that integrates large language models (LLMs) with a formal proof assistant, Lean. Prover Agent coordinates an informal reasoning LLM, a formal prover model, and feedback from Lean while also generating auxiliary lemmas. These auxiliary lemmas are not limited to subgoals in the formal proof but can also include special cases or potentially useful facts derived from the assumptions, which help in discovering a viable proof strategy. It achieves an 88.1% success rate on the MiniF2F benchmark, establishing a new state-of-the-art among methods using small language models (SLMs) with a much lower sample budget than previous approaches. We also present theoretical analyses and case studies that illustrate how these generated lemmas contribute to solving challenging problems. Our code is publicly available at: https://github.com/kAIto47802/Prover-Agent. |
| 18 | RvLLM: LLM Runtime Verification with Domain Knowledge | 2025-05-24 | 展开Large language models (LLMs) have emerged as a dominant AI paradigm due to their exceptional text understanding and generation capabilities. However, their tendency to generate inconsistent or erroneous outputs challenges their reliability, especially in high-stakes domains requiring accuracy and trustworthiness. Existing research primarily focuses on detecting and mitigating model misbehavior in general-purpose scenarios, often overlooking the potential of integrating domain-specific knowledge. In this work, we advance misbehavior detection by incorporating domain knowledge. The core idea is to design a general specification language that enables domain experts to customize domain-specific predicates in a lightweight and intuitive manner, supporting later runtime verification of LLM outputs. To achieve this, we design a novel specification language, ESL, and introduce a runtime verification framework, RvLLM, to validate LLM output against domain-specific constraints defined in ESL. We evaluate RvLLM on three representative tasks: violation detection against Singapore Rapid Transit Systems Act, numerical comparison, and inequality solving. Experimental results demonstrate that RvLLM effectively detects erroneous outputs across various LLMs in a lightweight and flexible manner. The results reveal that despite their impressive capabilities, LLMs remain prone to low-level errors due to limited interpretability and a lack of formal guarantees during inference, and our framework offers a potential long-term solution by leveraging expert domain knowledge to rigorously and efficiently verify LLM outputs. |
| 19 | APOLLO: Automated LLM and Lean Collaboration for Advanced Formal Reasoning | 2025-05-09 | 展开Formal reasoning and automated theorem proving constitute a challenging subfield of machine learning, in which machines are tasked with proving mathematical theorems using formal languages like Lean. A formal verification system can check whether a formal proof is correct or not almost instantaneously, but generating a completely correct formal proof with large language models (LLMs) remains a formidable task. The usual approach in the literature is to prompt the LLM many times (up to several thousands) until one of the generated proofs passes the verification system. In this work, we present APOLLO (Automated PrOof repair viaLLM and Lean cOllaboration), a modular, model-agnostic agentic framework that combines the strengths of the Lean compiler with an LLM's reasoning abilities to achieve better proof-generation results at a low token and sampling budgets. Apollo directs a fully automated process in which the LLM generates proofs for theorems, a set of agents analyze the proofs, fix the syntax errors, identify the mistakes in the proofs using Lean, isolate failing sub-lemmas, utilize automated solvers, and invoke an LLM on each remaining goal with a low top-K budget. The repaired sub-proofs are recombined and reverified, iterating up to a user-controlled maximum number of attempts. On the miniF2F benchmark, we establish a new state-of-the-art accuracy of 84.9% among sub 8B-parameter models (as of August 2025) while keeping the sampling budget below one hundred. Moreover, Apollo raises the state-of-the-art accuracy for Goedel-Prover-SFT to 65.6% while cutting sample complexity from 25,600 to a few hundred. General-purpose models (o3-mini, o4-mini) jump from 3-7% to over 40% accuracy. Our results demonstrate that targeted, compiler-guided repair of LLM outputs yields dramatic gains in both efficiency and correctness, suggesting a general paradigm for scalable automated theorem proving. |
| 20 | ShieldAgent: Shielding Agents via Verifiable Safety Policy Reasoning | 2025-03-26 | 展开Autonomous agents powered by foundation models have seen widespread adoption across various real-world applications. However, they remain highly vulnerable to malicious instructions and attacks, which can result in severe consequences such as privacy breaches and financial losses. More critically, existing guardrails for LLMs are not applicable due to the complex and dynamic nature of agents. To tackle these challenges, we propose ShieldAgent, the first guardrail agent designed to enforce explicit safety policy compliance for the action trajectory of other protected agents through logical reasoning. Specifically, ShieldAgent first constructs a safety policy model by extracting verifiable rules from policy documents and structuring them into a set of action-based probabilistic rule circuits. Given the action trajectory of the protected agent, ShieldAgent retrieves relevant rule circuits and generates a shielding plan, leveraging its comprehensive tool library and executable code for formal verification. In addition, given the lack of guardrail benchmarks for agents, we introduce ShieldAgent-Bench, a dataset with 3K safety-related pairs of agent instructions and action trajectories, collected via SOTA attacks across 6 web environments and 7 risk categories. Experiments show that ShieldAgent achieves SOTA on ShieldAgent-Bench and three existing benchmarks, outperforming prior methods by 11.3% on average with a high recall of 90.1%. Additionally, ShieldAgent reduces API queries by 64.7% and inference time by 58.2%, demonstrating its high precision and efficiency in safeguarding agents. |
| 21 | Nonlocal perimeters and variations: Extremality and decomposability for finite and infinite horizons | 2025-02-07 | 展开We analyze the extremality and decomposability properties with respect to two types of nonlocal perimeters available in the literature, the Gagliardo perimeter based on the eponymous seminorms and the nonlocal distributional Caccioppoli perimeter, both with finite and infinite interaction ranges. A nonlocal notion of indecomposability associated to these perimeters is introduced, and we prove that in both cases it can be characterized solely in terms of the interaction range or horizon |
感谢原始项目 @zezhishao/DailyArXiv 提供的灵感。