Yi Wang

InternVideo-Next: Towards World-Understanding Video Models

VideoChat-Flash: Hierarchical Compression for Long-Context Video Modeling

UniFlow: A Unified Pixel Flow Tokenizer for Visual Understanding and Generation

RIVER: Real-time Video Interaction Benchmark

Balancing the Experts: Unlocking LoRA-MoE for GRPO via Mechanism-Aware Rewards

VideoChat-r1.5: visual test-time scaling to reinforce multimodal reasoning by iterative perception

VRBench: a benchmark for multi-step reasoning in long narrative videos

Make your training flexible: towards deployment-efficient video models

TimeSuite: Improving MLLMs for Long Video Understanding via Grounded Tuning