Junyu Zhang

I am a Master's student in Computer Science (MSCS) at University of Illinois Urbana-Champaign, advised by Prof. Huan Zhang. Previously, I received my bachelor's degree from Huazhong University of Science and Technology. I am fortunate to work under the supervision of Prof. Paul Pu Liang at MIT, Prof. Chongjie Zhang in the Institute for Interdisciplinary Information Sciences at Tsinghua University and Prof. Chuang Gan in the MIT-IBM Watson AI Lab.

In the era of large-scale foundation models, my research goal is to incentivize the capabilities of pretrained models by empowering them to perform long-term reasoning and learn from interaction, thereby building scalable and generalizable AI for the open world. Toward this goal, my previous research primarily focused on

• Large Language Models Reasoning: efficient and theoretically grounded reasoning; embodied spatial reasoning for agentic decision-making; mathematical reasoning with scalable evaluation.
• Learning from Interactions: empirical RL algorithms and their applications in robotics, including morphology–behavior co-optimization and robotic manipulation.

If you would like to discuss my research or potential collaborations further, feel free to contact me. I'm open to connect and collaborate with each other.

Email  /  CV (last updated: Oct 2025)  /  Google Scholar  /  Github  /  Twitter

AlphaOne: Reasoning Models Thinking Slow and Fast at Test Time

EmbodiedBench: Comprehensive Benchmarking Multi-modal Large Language Models for Vision-Driven Embodied Agents

DynaMath: A Dynamic Visual Benchmark for Evaluating Mathematical Reasoning Robustness of Vision Language Models

Leveraging Hyperbolic Embeddings for Coarse-to-Fine Robot Design

Symmetry-Aware Robot Design with Structured Subgroups

Offline Meta Reinforcement Learning with In-Distribution Online Adaptation