AI Paper Atlas

Research map / Video / Video-language understanding

OmniZip: Audio-Guided Dynamic Token Compression for Fast Omnimodal Large Language Models

Keda Tao, Kele Shao, Bohan Yu, Weiqiang Wang, Jian Liu, Huan Wang

CVPR 2026 · Video › Video-language understanding · approach: Video LLMs with fine-grained understanding · 25 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add audio-guided dynamic video pruning allocation (ablation)OmniZip · WorldSense (45% retained)avg score: 45.0 -> 45.9Improved
add audio anchor consolidation (ablation)OmniZip · WorldSense (45% retained)avg score: 44.8 -> 45.0Improved
use ISTC instead of DyCoke for video token compression (ablation)OmniZip · WorldSense (45% retained)avg score: 45.0 -> 45.9Improved
use window-wise selection instead of VisionZip global selection (ablation)OmniZip · WorldSense (45% retained)avg score: 44.2 -> 45.9Improved
use OmniZip attention-based audio selection instead of random audio selection (ablation)OmniZip with ISTC · WorldSense (45% retained)avg score: 45.1 -> 45.9Improved
apply OmniZip at 35% retention versus FastV and DyCoke at 50% retentionQwen2.5-Omni-7B · WorldSenseavg score: 44.3 / 44.6 -> 45.3Improved
prune audio or video tokens too aggressively (ablation)OmniZip · WorldSense sensitivity to rho_a and rho_vavg scoreHurt
apply FastV-style attention-based pruning to omnimodal tokensQwen2.5-Omni-7B · AVUT / WorldSenseaccuracy: FastV 94.3% vs OmniZip 99.1% of full at 45-50% retentionHurt
use smaller model (3B vs 7B) under token compressionOmniZip · Omni QA averagerelative performance: 99.1% (7B) vs 99.2% (3B) at 45% retentionNo clear effect

Later papers that compared with this method

4 later accepted papers report OmniZip as a baseline in their results tables; 4 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy