Limin Wang

6 papers Nanjing University

• Eagle 2.5: Boosting Long-Context Post-Training for Frontier Vision-Language Models
• LongVPO: From Anchored Cues to Self-Reasoning for Long-Form Video Preference Optimization
• MUVR: A Multi-Modal Untrimmed Video Retrieval Benchmark with Multi-Level Visual Correspondence
• MotionRAG: Motion Retrieval-Augmented Image-to-Video Generation
• StreamForest: Efficient Online Video Understanding with Persistent Event Memory
• VideoChat-R1.5: Visual Test-Time Scaling to Reinforce Multimodal Reasoning by Iterative Perception