AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Exploring Mode Connectivity in Krylov Subspace for Domain Generalization

Aodi Li, Liansheng Zhuang, Xiao Long, Houqiang Li, Shafei Wang

ICLR 2026 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Optimization-based domain generalization

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 Krylov-inspired initial direction instead of random initial direction (ablation)BOA · VLCS LabelMe, loss variation 0.3test accuracy: 69.2 -> 70.8; 0.5-1.6% across all loss variation valuesImproved
add Krylov-subspace-constrained reflection on top of Krylov-inspired p0 (ablation)BOA · VLCS LabelMe, loss variation 0.3test accuracy: 70.8 -> 75.7Improved
full BOA within Krylov subspace versus random p0 (ablation)BOA · VLCS LabelMe, loss variation 0.3test accuracy: 69.2 -> 75.7Improved
add BOA on top of SAM with ViT-B/16 prompt tuningSAM · DomainBed average (ViT-B/16)OOD accuracy: 76.2 -> 78.1Improved
apply BOA instead of SAM with ViT-B/32ViT-B/32 CLIP VPT · DomainBed average (ViT-B/32)OOD accuracy: 70.9 -> 73.3Improved
apply BOA instead of SAM with ViT-L/14ViT-L/14 CLIP VPT · DomainBed average (ViT-L/14)OOD accuracy: 80.1 -> 81.5Improved
apply BOA instead of SAM with ResNet50ResNet50 · VLCS (ResNet50)OOD accuracy: 80.2 -> 81.8Improved
use oracle negative test gradient as BOA initial direction (ablation)ERM and SAM warm-up · DomainBed average (ViT-B/16)OOD accuracy: 75.5 -> 80.4 (ERM); 76.2 -> 81.0 (SAM)Improved
move along approximate Krylov initial direction (beta_k=1) without reflection (ablation)SAM warm-up model on VLCS · VLCS domainstest accuracy: +1.4% Caltech101, +6.5% LabelMe, +2.7% SUN09, +2.9% VOC2007Improved
use Krylov subspace instead of random subspace for approximating test gradient (ablation)ResNet50 gradient alignment · SUN09 cosine similarity, K=10cosine similarity: 0.2657 -> 0.7617Improved
use Krylov subspace instead of random subspace on Caltech101 (ablation)ResNet50 gradient alignment · Caltech101 cosine similarity, K=10cosine similarity: 0.3125 -> 0.3381No clear effect
increase Krylov subspace dimension from 1 to 7-10 (ablation)test gradient alignment on LabelMe (ViT) · LabelMe cosine similaritycosine similarity: cos < 0.1 at K=1 -> above 0.8 at K=7-10Improved

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