AI Paper Atlas

Research map / Video / Video-language understanding

Sink-Token-Aware Pruning for Fine-Grained Video Understanding in Efficient Video LLMs

Kibum Kim, Jiwan Kim, Kyle Min, Yueqi Wang, Jinyoung Moon, Julian McAuley, Chanyoung Park

ECCV 2026 · Video › Video-language understanding · approach: Long video understanding · 2 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 sink-token-aware spatial pruning (STSP) (ablation)VisionZip · fine-grained suite, 10% retentionperformance drop rate: 20.46% -> 4.64%Improved
add STSP (ablation)Holitom · fine-grained suite, 10% retentionperformance drop rate: 3.87% -> 1.94%Improved
add sink-token-aware temporal pruning (STTP) on top of STSP (ablation)Holitom · fine-grained suite, 10% retentionperformance drop rate: 1.94% -> 1.17%Improved
add SToPFastVid · fine-grained suite, 10% retentionperformance drop rate: 15.69% -> 6.32%Improved
add SToPVisionZip · EventHallusion Binary, 10% retentionaccuracy: 50.37 -> 60.39Improved
add SToPHolitom · fine-grained suite, 20% retentionperformance drop rate: 2.41% -> 1.84%Improved
add SToPVisionZip on LLaVA-Video-7B · fine-grained suite, 10% retentionperformance drop rate: 26.46% -> 7.04%Improved
add SToPVisionZip, FastVid, Holitom · MCQA suite, 10% retentionperformance drop rate: 7.36%->3.34%, 5.12%->2.90%, 3.21%->2.02%Improved
remove temporal pruning (ablation)Holitom · EventHallusionhallucination performanceHurt
remove sink tokens at increasing ratio from selected set (ablation)VisionZip · EventHallusion and MCQAperformance degradationImproved
use attention-based pruning + SToP instead of feature-based DPC-KNN pruning (ablation)spatial pruning · fine-grained suite, 10% retentionperformance drop rate: 6.50% -> 4.41%Improved
use feature-based pruning instead of attention-based pruning (ablation)spatial pruning · fine-grained suite, 10% retentionperformance drop rate: 15.85% -> 6.50%Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

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