AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Instruction-Guided Cross-Modal Clustering for Training-Free Visual Token Pruning in Vision-Language Models

Yunqian Yu, Biao Chen, Yunya Zhang, Tonglan Xie, Mengmeng Jing, Lin Zuo

AAAI 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Attention-based visual token pruning · 2 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes a training-free visual token pruning method for vision-language models that uses instruction-guided cross-modal clustering to remove redundant visual tokens and cut inference cost.

Problem

LVLMs have far more visual tokens than text tokens, causing redundancy and high inference cost; existing pruning relies on unimodal information or cross-modal attention, which have drawbacks.

Approach

Combines instruction (text) guidance with clustering over visual tokens in a cross-modal space to select representative visual tokens for pruning, without any training.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add CLS-attention token selection over random selection (ablation)ICCTP pruning (LLaVA-1.5-7B, 88.9% pruning) · TextVQAaccuracy: +5.0%Improved
add instruction-noun clustering on top of CLS-only selection (ablation)ICCTP pruning (LLaVA-1.5-7B, 88.9% pruning) · POPEF1: +5.0%Improved
add instruction-noun clustering on top of CLS-only selection (ablation)ICCTP pruning (LLaVA-1.5-7B, 88.9% pruning) · MMEscore: +65.6Improved
vary global retention ratio rho_g among 0.4, 0.6, 0.8 (ablation)ICCTP (LLaVA-1.5-7B, 88.9% pruning) · ten benchmarksaccuracyNo clear effect
vary semantic-context factor alpha among 0.2, 0.3, 0.4 (ablation)ICCTP (LLaVA-1.5-7B, 88.9% pruning) · ten benchmarksaccuracyNo clear effect
ICCTP vs VisionZip at 94.4% pruningLLaVA-1.5-7B · POPEaccuracy: 68.7 -> 80.4Improved
ICCTP vs VisionZip at 94.4% pruningLLaVA-NeXT-7B · TextVQAaccuracy: 54.7 -> 45.1Hurt
ICCTP vs DART at 94.4% pruningLLaVA-1.5-7B · MM-Vet / MMBench-CNscore: MMBCN 47.1 vs 50.0; MMVet 22.2 vs 25.0Mixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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