AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Structural Pruning of Large Vision Language Models: A Comprehensive Study on Pruning Dynamics, Recovery, and Data Efficiency

Yiran Huang, Lukas Thede, Massimiliano Mancini, Wenjia Xu, Zeynep Akata

IJCV 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. 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 widthwise instead of layerwise pruning (zero-shot) (ablation)LLaVA-v1.5-7B at 15% pruning · 9-benchmark averageAVG-%: 72.22% -> 90.87%Improved
use widthwise instead of layerwise pruning (zero-shot) (ablation)Bunny-v1-3B at 15% pruning · AI2Daccuracy: 52.73 -> 65.64Improved
use widthwise instead of layerwise pruning on free-form tasks (ablation)Bunny-v1-3B at 15% pruning · GQA, DocVQA, MathVistarelative performance: 2.2% -> 31.5%Improved
use layerwise instead of widthwise pruning after recovery training (ablation)pruned LVLMs at <=30% ratio · 9-benchmark averageAVGImproved
use widthwise instead of layerwise pruning after recovery training (ablation)pruned LVLMs at >=45% ratio · 9-benchmark averageAVGImproved
use layerwise instead of widthwise pruning for latency (ablation)LLaVA at 30% compression · inference latencylatency (ms): 84.4 -> 80.7Improved
use layerwise instead of widthwise pruning for FLOPs (ablation)LLaVA at 30% compression · FLOPsFLOPs (T): 6.89 -> 6.92No clear effect
compare multimodal vs text-only degradation at mild pruningLLaVA-v1.5-7B widthwise 15% · multimodal vs text-only inputsrelative degradation: 9.13% -> 1.07%Mixed
finetune only the multimodal projector (ablation)pruned LVLMs at 15% ratio · 9-benchmark averageAVG-%: over 95%Improved
additionally finetune the pruned LLM with LoRA (ablation)pruned LVLMs at 45% ratio · 9-benchmark averageAVG-%: over 80%Improved
replace SFT with standalone KD (KL, RKL or L2) (ablation)pruned LVLMs at high ratios · 9-benchmark averageAVGHurt
add L2 hidden-state distillation to SFT (ablation)pruned LVLMs · 9-benchmark averageAVGImproved

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