Chi (Alan) Wan

Chi Wan δΈ‡ι©°

AI Researcher

πŸ‘‹πŸ‘‹Hi, I'm Chi! I am currently working full-time with Prof. Yiming Li on some crazy and interesting works. I graduated from Northwestern University with M.S. in Computer Engineering, supervised by Prof. Manling Li, and collaborated with Prof. Huajie Shao. Previously, I worked in the Machine Vision and Robotics Laboratory at Northeastern University, mentored by Prof. Kechen Song, where I obtained my bachelor's degree in Intelligent Manufacturing Engineering.

πŸ“Š Data-Driven Large Scale Training 🌍 World Models with Physical Laws πŸ€– Embodied AI
Chi Wan

Research Focus

Data Collection & Curation

I believe high-quality, carefully curated data is becoming a foundational advantage for embodied agents.

Physics-Grounded World Models

I am interested in world models that learn through physical-world interaction and reflect the constraints of real-world physics.

Embodied Agents

I aim to develop embodied agents capable of continuous adaptation, self-improvement, and lifelong learning.

News

: I graduated from Northwestern University and started working with Prof. Yiming Li.

: WorldAgen: Unified State-Action Prediction with Test-Time World Model Training accepted to AAAI 2026.

: VAGEN: Reinforcing World Model Reasoning for Multi-Turn VLM Agents accepted to NeurIPS 2025.

Publications

* equal contribution

WorldAgen project preview

WorldAgen: Unified State-Action Prediction with Test-Time World Model Training

Chi Wan*, Kangrui Wang*, Yuan Si, Pingyue Zhang, Huang Huang, Manling Li

AAAI 2026

VAGEN project preview

VAGEN: Reinforcing World Model Reasoning for Multi-Turn VLM Agents

Kangrui Wang*, Pingyue Zhang*, Zihan Wang*, Yaning Gao*, Linjie Li*, Qineng Wang, Hanyang Chen, Chi Wan, Yiping Lu, Zhengyuan Yang, Lijuan Wang, Ranjay Krishna, Jiajun Wu, Li Fei-Fei, Yejin Choi, Manling Li

NeurIPS 2025 · 400+ GitHub stars🌟

LiDAR scanning device preview

Design and Development of a High-Resolution Panoramic LiDAR Scanning Device

Chi Wan

Northeastern University Graduation Thesis

TSSTNet project preview

TSSTNet: A Two-Stream Swin Transformer Network for Salient Object Detection of No-Service Rail Surface Defects

Chi Wan, Shuai Ma and Kechen Song

Coatings

Education

Northwestern University

Master of Science in Computer Engineering Β· GPA: 3.97/4.0

Advisor: Prof. Manling Li

Northeastern University

Bachelor of Science in Intelligent Manufacturing Engineering Β· GPA: 87.6/100

Advisor: Prof. Kechen Song

Services

Conference Technical Program Reviewer

AAAI 2026 Β· NeurIPS 2025 Β· ICLR 2026