GitHub·Code·MIT

Fanziyang-v/FlashVID

[ICLR 2026 Oral] FlashVID: Efficient Video Large Language Models via Training-free Tree-based Spatiotemporal Token Merging

115 stars·Updated Jul 2026
efficiencyflashvidmultimodaltoken-compressionvideo-llms
Open on GitHub

Authors

Ziyang FanSouthwest University
Repository owner

Linked papers