About Me

Hello! I am a senior undergraduate student at HUST and an incoming PhD student at SJTU, where I am advised by Prof. Zhijie Deng.

The way humans interact with the physical world is generally a continuous cycle of cognition, action, and perception. I believe that to achieve human-like intelligence, we must build a system with a similar closed-loop mechanism. Such a system would continuously learn and evolve through ongoing interactions with environment, developing the capacity to perceive, understand, reason, and create within the physical world.

Motivated by this curiosity, my research focuses on developing machine learning methodologies for generative modeling and reinforcement learning, particularly in multimodal generation and embodied AI.

Selected Publications

LightningRL framework showing policy optimization, sampling and reward, and block-wise diffusion language model inference before and after training.

LightningRL: Breaking the Accuracy-Parallelism Trade-off of Block-wise dLLMs via Reinforcement Learning

International Conference on Machine Learning (ICML), Seoul, Korea, 2026

Yanzhe Hu, Yijie Jin, Pengfei Liu, Kai Yu, Zhijie Deng

Representative dependency-tree topologies and interconnection statistics from the large language model software supply-chain study.

Understanding Large Language Model Supply Chain: Structure, Domain, and Vulnerabilities

Internetware, 2026

Yanzhe Hu*, Shenao Wang*, Tianyuan Nie, Yanjie Zhao, Haoyu Wang