AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

Escaping Optimization Stagnation: Taking Steps Beyond Task Arithmetic via Difference Vectors

Jinping Wang, Zhiqiang Gao, Zhang Dinggen, Zhiwu Xie

AAAI 2026 · Architectures and efficiency › Pruning and efficient tuning · approach: Parameter-efficient transfer analysis · 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 difference vector direction instead of random perturbation of same magnitude (ablation)DV-BASI · Unsupervised model merging (ViT-B/32)accuracy: 5.6 -> 83.9Improved
use anisotropic block-wise scaling instead of scalar scaling (ablation)DV-BASI · Unsupervised model merging (ViT-B/16)accuracy: 83.6 -> 85.3Improved
add DV-BASI to aTLASaTLAS · Task addition, 8 datasets (unsupervised)absolute accuracy: 82.8 -> 85.3 (ViT-B/16); 90.2 -> 90.8 (ViT-L/14)Improved
add DV-BASI to tauJP regtauJP reg · Task addition, 8 datasets (unsupervised)absolute accuracy: 84.5 -> 86.7 (ViT-B/32)Improved
add DV-BASI to supervised aTLASaTLAS · Task addition, 8 datasets (supervised)absolute accuracy: 82.9 -> 87.2 (ViT-B/16)Improved
add multi-objective optimization (MOO) to DV-BASI (ablation)aTLAS + DV-BASI · Task addition, 8 datasets (supervised)absolute accuracy: 85.9 -> 86.2 (ViT-B/32)Improved
add DV-BASI to aTLAS for task negationaTLAS · Task negation target accuracy (supervised)target accuracy: 19.4 -> 10.7 (ViT-B/32)Improved
add DV-BASI to tauJP reg for task negation (unsupervised)tauJP reg · Task negation target accuracy (unsupervised)target accuracy: 6.7 -> 5.7 (ViT-B/32)Improved
add DV-BASI to aTLAS for test-time adaptationaTLAS · Test-time adaptation (22 datasets)accuracy: 65.16 -> 66.76 (ViT-B/32)Improved
apply DV-BASI after single-task fine-tuningfine-tuned CLIP ViT-B/32 · 20 datasets averageaccuracy: 82.04 -> 82.74Improved
vary initial scaling coefficient from 0.1 to 1.0 (ablation)DV-BASI · Unsupervised model mergingaccuracyNo clear effect

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