Large Model Innovation Center
Open Menu
Close Menu
Home
News
Research
Publication
English
中文 (简体)
Xiangyu Zeng
VideoChat-Flash: Hierarchical Compression for Long-Context Video Modeling
Apr 24, 2026
UniFlow: A Unified Pixel Flow Tokenizer for Visual Understanding and Generation
Apr 24, 2026
RIVER: Real-time Video Interaction Benchmark
Apr 24, 2026
Balancing the Experts: Unlocking LoRA-MoE for GRPO via Mechanism-Aware Rewards
Apr 24, 2026
VideoChat-r1.5: visual test-time scaling to reinforce multimodal reasoning by iterative perception
Oct 11, 2025
StreamForest: efficient online video understanding with persistent event memory
Oct 11, 2025
Make your training flexible: towards deployment-efficient video models
Aug 12, 2025
Task preference optimization: improving multimodal large language models with vision task alignment
Apr 20, 2025
Online video understanding: a comprehensive benchmark and memory-augmented method
Apr 20, 2025
TimeSuite: Improving MLLMs for Long Video Understanding via Grounded Tuning
Apr 15, 2025