ZhijunLStudio

`$ cat /about.txt`

I'm a researcher focused on making LLMs faster and smarter. Currently working on inference optimization, multimodal post-training (SFT/RLHF/DPO/GRPO), and efficient serving at scale.

Previously explored computer vision, now deep into the world of large language models and reinforcement learning.

`$ cat /interests.txt`

Inference Optimization (Quantization, KV-Cache, Speculative Decoding)

Post-training Alignment (SFT, RLHF, DPO, GRPO)

Multimodal LLMs (Vision-Language, Audio-Language)

Distributed Training & Efficient Serving (vLLM, TensorRT-LLM)