Research archive

Publications

* Equal contribution · † Corresponding author

First-page preview of When Prompts Become Pixels

2026 · Preprint

When Prompts Become Pixels: Prompt-Region Grounding for Multimodal Reasoning

Yongxin Wang^*, Ruizhe Zhou^*, Yueling Tang, Yingying Zhu, Xuemin Zhao, Xiaojun Chang, Xiaodan Liang^†

arXiv · 2026

TL;DR A prompt-region grounding method that connects textual prompts to image regions, enabling multimodal models to reason with spatially grounded visual evidence.

Paper

First-page preview of A Survey on Physical AI

2026 · Survey

Aligning Perception, Reasoning, Modeling and Interaction: A Survey on Physical AI

Kun Xiang^*, Terry Jingchen Zhang^*, Yinya Huang^*, Jixi He, Zirong Liu, Yueling Tang, Ruizhe Zhou, Lijing Luo, Youpeng Wen, Xiuwei Chen, Bingqian Lin, Jianhua Han, Hang Xu, Hanhui Li, Bin Dong, Xiaodan Liang^†

T-PAMI · 2026

TL;DR A survey of Physical AI that unifies perception, reasoning, world modeling, and interaction into a framework for intelligent systems acting in the physical world.

Paper Project

First-page preview of Models Under SCOPE

2026 · Method

Models Under SCOPE: Scalable and Controllable Routing via Pre-hoc Reasoning

Qi Cao^*, Shuhao Zhang^*, Ruizhe Zhou, Ruiyi Zhang, Peijia Qin, Pengtao Xie^†

ICML · 2026

TL;DR A scalable, controllable model-routing framework that uses pre-hoc reasoning to select an appropriate model before generating a response.

Paper Code Project

2025 · Position paper

Position: The Physics-Physical Reasoning Interplay is Key for Future Embodied World Models

Terry Jingchen Zhang^*, Kun Xiang^*, Yinya Huang^*, Jixi He, Zirong Liu, Yueling Tang, Ruizhe Zhou, Chengyu Yu, Xiaodan Liang^†

NeurIPS LAW Workshop · 2025

TL;DR A position paper arguing that embodied world models need both knowledge of physics and physical reasoning to predict and interact with the world reliably.

Paper