Yan Lu

9 papers The Chinese University of Hong Kong

• Deep Video Discovery: Agentic Search with Tool Use for Long-form Video Understanding
• FuncGenFoil: Airfoil Generation and Editing Model in Function Space
• Image as a World: Generating Interactive World from Single Image via Panoramic Video Generation
• Omnidirectional 3D Scene Reconstruction from Single Image
• One-Step Diffusion-Based Image Compression with Semantic Distillation
• PRING: Rethinking Protein-Protein Interaction Prediction from Pairs to Graphs
• STAR: A Benchmark for Astronomical Star Fields Super-Resolution
• Scientists' First Exam: Probing Cognitive Abilities of MLLM via Perception, Understanding, and Reasoning
• VideoVLA: Video Generators Can Be Generalizable Robot Manipulators