Ziyang Fan — accepted AI-conference papers

2026

FlashVID: Efficient Video Large Language Models via Training-free Tree-based Spatiotemporal Token Merging

Ziyang Fan, Keyu Chen, Ruilong Xing, Yulin Li, Li Jiang, Zhuotao Tian

ICLR 2026oral

Although Video Large Language Models (VLLMs) have shown remarkable capabilities in video understanding, they are required to process high volumes of visual tokens, causing significant computational inefficiency. Existing VLLMs acceleration frameworks usually compress spatial and temporal redundancy…

Cited by 0 Source ↗code ↗Scholar ↗

2025

Less Is More, but Where? Dynamic Token Compression via LLM-Guided Keyframe Prior

Yulin Li, Haokun GUI, Ziyang Fan, Junjie Wang, Bin Kang, BIN CHEN +1

NeurIPS 2025poster

Recent advances in Video Large Language Models (VLLMs) have achieved remarkable video understanding capabilities, yet face critical efficiency bottlenecks due to quadratic computational growth with lengthy visual token sequences of long videos. While existing keyframe sampling methods can improve te…

Cited by 0 Source ↗Scholar ↗