news

Mar 01, 2026 Two papers accepted at CVPR 2026: VQ-VA World and An Empirical Study on How Video-LLMs Answer Video Questions.
Oct 01, 2025 Joined ByteDance Seed Edge as a research intern, working on Agentic and Self-Evolving AI.
Jun 01, 2025 Two papers accepted at CVPR 2025: DrVideo and Point-Cache.
Jan 01, 2025 BAGEL (Emerging Properties in Unified Multimodal Pretraining) released, reaching 5.5k GitHub stars.
Jun 01, 2024 Co-organized the CVPR 2024 Workshop on Robot Visual Perception in Human Crowded Environments and the JRDB-PanoTrack Challenge.
Sep 15, 2022 RTFormer accepted as a Spotlight at NeurIPS 2022.