AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

LoGIC: Multi-LoRA Guided Importance Consensus for Multi-Task Pruning in Vision Transformers

Yu-Hong Chou, Rui Fang, Hsi-Wen Chen, Ming-Syan Chen

AAAI 2026 · Architectures and efficiency › Attention and transformers · approach: Vision transformer pruning · 0 citations (OpenAlex)

Read the paper

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
LoGIC instead of best competing MTT pruning methodViT-L, 30% sparsity · Five-task averageaverage score: 0.6037 -> 0.6943Improved
LoGIC vs LoRAPrune on FFT backbonesViT-L, 50% sparsity · NYUv2 segmentationmIoU: 0.6540 -> 0.7130Improved
LoGIC vs MTT + Magnitude PruningViT-L, 50% sparsity · Oxford Flowers-102accuracy: 0.5581 -> 0.9006Improved
LoGIC vs Swin-L LoRAPrune at 50% sparsitySwin-L, 50% sparsity · NYUv2 segmentationmIoU: 0.4782 -> 0.6201Improved
30% LoGIC vs dense LoRA on ViT-LLoGIC ViT-L 30% · NYUv2 segmentationmIoU: 0.7531 -> 0.7506Hurt
LoGIC vs dense FFT on detection at 50% sparsityViT-L, 50% sparsity · CPPE-5mAP: 0.4457 -> 0.4400Hurt
Pruning 30% to 50% with LoGICViT-L · Five-task averageaverage score: 0.6943 -> 0.6580Hurt
LoGIC vs full FFT in parametersViT-L · Deployment sizetotal parameters: 1520M -> 243.6MImproved

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