Papers

VideoPhy-2: A Challenging Action-Centric Physical Commonsense Evaluation in Video Generation

Hritik Bansal, Clark Peng, Yonatan Bitton, Roman Goldenberg, Aditya Grover, Kai-Wei Chang

ICLR 2026; Best Paper at ICML 2025 World Models Workshop

A benchmark for whether generated videos obey everyday physical cause and effect.

DialectGen: Benchmarking and Improving Dialect Robustness in Multimodal Generation

Yu Zhou, Sohyun An, Haikang Deng, Da Yin, Clark Peng, Cho-Jui Hsieh, Kai-Wei Chang, Nanyun Peng

ResponsibleFM @ NeurIPS 2025

A study of how image and video models handle English dialects, plus ways to make them fairer.

Event Detection via Probability Density Function Regression

Clark Peng, Tolga Dincer

arXiv 2024

A way to find event boundaries by predicting where changes are most likely to happen.