Mike Zheng Shou

8 papers National University of Singapore

• CoFFT: Chain of Foresight-Focus Thought for Visual Language Models
• DOTA: Distributional Test-time Adaptation of Vision-Language Models
• OmniConsistency: Learning Style-Agnostic Consistency from Paired Stylization Data
• PANDA: Towards Generalist Video Anomaly Detection via Agentic AI Engineer
• Show-o2: Improved Native Unified Multimodal Models
• Sparse Image Synthesis via Joint Latent and RoI Flow
• Think or Not? Selective Reasoning via Reinforcement Learning for Vision-Language Models
• macOSWorld: A Multilingual Interactive Benchmark for GUI Agents