Token Reduction via Local and Global Contexts Optimization for Efficient Video Large Language Models
Open paper
Benchmark
Model
Rank
Results
zero-shot-video-question-answer-on-egoschema-fullset
AOT (LLaVA-OneVision-7B)
#26
Accuracy: 61.7
video-question-answering-on-mvbench
AOT (LLaVA-Video-7B)
#73
Avg.: 59.2
Rank counts only results with a code link.