I am a third-year master’s student at the Institute of Automation, Chinese Academy of Sciences, under the supervision of Prof. Haifeng Zhang , collaborated with Prof. Jun Wang . My research focuses on Hardware Co-design for LLM, Agentic RL with tool use, and efficient LLM inference.
In collaboration with Li Auto, we explored a hardware co-design law that jointly characterizes model accuracy and inference performance ( arXiv:2602.10377 ).
I have been curating resources for CUDA programming: CUDA .
I have been curating a list of models for code generation: LLMs for Code .
I have updated a curated list of resources for reinforcement learning on GitHub , I hope it helps if you are studying RL as well.
AAMAS 2026 The 25th International Conference on Autonomous Agents and Multiagent Systems
We introduce Feasible Constraint Policy Optimization (FCPO), which seamlessly combines penalty and trust region methods to address policy feasibility while ensuring stability and performance.
arXiv 2025 arXiv preprint
We introduce the PLM, a Peripheral Language Model, developed through a co-design process that jointly optimizes model architecture and edge system constraints.
Li Auto, MindGPT & MindVLA
Meituan, RL Infra & Algorithms
AMD, LLM & MLLM Inference Acceleration
