Homepage

Jiajian Li 李佳键

PhD Student

MoE Key Lab of Artificial Intelligence, AI Institute, School of Computer Science, Shanghai Jiao Tong University

About

Biography

I am a PhD student at Shanghai Jiao Tong University, advised by Yunbo Wang. My research interests include world models, reinforcement learning, and embodied AI.

I am broadly interested in building physically grounded models that connect perception, dynamics, and control for intelligent agents, with a particular focus on simulation and embodied decision making.

Publications

Selected Works

Overview figure for OrbiSim

World Model / Embodied AI

OrbiSim: World Models as Differentiable Physics Engines for Embodied Intelligence

Jiajian Li^*, Jingyuan Huang^*, Junru Gong^*, Qi Wang, Xiaokang Yang, Yunbo Wang^†

arXiv

PDF Website Code

Overview figure for VeoRL

Reinforcement Learning / World Model

Video-Enhanced Offline Reinforcement Learning: A Model-Based Approach

Minting Pan, Yitao Zheng, Jiajian Li, Yunbo Wang^†, Xiaokang Yang

ICML 2025

PDF Website Code

Overview figure for LS-Imagine

World Model / Reinforcement Learning

Open-World Reinforcement Learning over Long Short-Term Imagination

Jiajian Li^*, Qi Wang^*, Yunbo Wang^†, Xin Jin, Yang Li, Wenjun Zeng, Xiaokang Yang

ICLR 2025 Oral (Top 1.8%)

PDF Website Code