Dimensions overall score 7.0

GitHub Code Pulse

No public code linked for this paper yet.

• HieraVid achieves new state-of-the-art performance on video understanding benchmarks while retaining only 30% of tokensresult 95%
• HieraVid maintains over 98% of the performance of LLaVA-Video-7Bresult 95%
• HieraVid maintains over 99% of the performance of LLaVA-OneVision-7Bresult 95%
• HieraVid uses a hierarchical pruning framework with three levels: segment-level, frame-level, and layer-levelmethod 90%
• Existing video token pruning methods mainly prune at input level while neglecting inherent information structurelimitation 85%
• HieraVid progressively and dynamically reduces visual redundancy in videosmethod 85%
• The method is based on observations that videos possess segment-frame structure and LLMs propagate multi-modal information unidirectionallytechnical 80%
• Extensive experiments were conducted on four widely used video understanding benchmarksresult 90%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card