AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

AnchorPrune: Relevance-Anchored Contextual Expansion for Visual Token Pruning

Kyuan Oh, Bumsoo Kim

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Attention-based visual token pruning · 1 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
use importance-weighted contextual expansion instead of diversity-only selection in Stage 2 (ablation)AnchorPrune · LLaVA-1.5-7B, 64 tokensRel.: 85.4 -> 91.0Improved
use importance-weighted contextual expansion instead of the best alternative coupling at 32 tokens (ablation)AnchorPrune · LLaVA-1.5-7B, 32 tokensRel.: 81.1 -> 85.1Improved
use additive relevance-diversity coupling instead of diversity-only (ablation)AnchorPrune Stage 2 · LLaVA-1.5-7B, 64 tokensRel.: 85.4 -> 87.3Improved
use negated instead of raw CLIP patch-text similarity in Stage 1 (ablation)AnchorPrune · LLaVA-1.5-7B, 32 tokensMME: 1245.4 -> 1394.7Improved
use negated instead of raw CLIP similarity in Stage 1 (POPE) (ablation)AnchorPrune · LLaVA-1.5-7B, 32 tokensPOPE: 74.2 -> 85.3Improved
use negated instead of raw CLIP similarity in Stage 1 (TextVQA) (ablation)AnchorPrune · LLaVA-1.5-7B, 32 tokensTextVQA: 53.7 -> 54.2No clear effect
use token-wise max matching instead of mean aggregation of instruction embeddings (ablation)AnchorPrune · Qwen2.5-VL-7B, 256 tokensMME: 1647.4 -> 1657.5Improved
restore retained tokens to raster-scan order instead of raw selection order (ablation)AnchorPrune · LLaVA-1.5-7B, 64 tokensMME-OCR: 132.5 -> 140.0Improved
restore raster-scan order (OCRBench) (ablation)AnchorPrune · LLaVA-1.5-7B, 64 tokensOCRBench: 26.9 -> 28.9Improved
set patience P=3 instead of P=1 (ablation)AnchorPrune · LLaVA-1.5-7B, 32 tokensMME: 1384.8 -> 1394.7Improved
set patience P=3 instead of P=5 (ablation)AnchorPrune · Qwen2.5-VL-7B, 64 tokensMME: 1552.2 -> 1563.2Improved
vary novelty threshold tau from 0.1 to 0.4 (ablation)AnchorPrune · LLaVA-1.5-7B, 32 tokensMME: 1388.9 to 1394.7No clear effect

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