AI Paper Atlas

Research map / Label-efficient and robust learning / Test-time adaptation

CS-TTA: Preserving Concept Sensitivity in Test-Time Adaptation

Junah Jung, YeonGyu Han, Chang Min Park, Dongheon Lee

ECCV 2026 · Label-efficient and robust learning › Test-time adaptation · approach: Pseudo-label test-time adaptation · 0 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
add CS-TTA regularizerTent · Waterbirdsworst-group accuracy: 56.2 -> 65.8Improved
add CS-TTA regularizerDeYO · Waterbirdsworst-group accuracy: 63.8 -> 68.5Improved
add CS-TTA regularizerfive TTA baselines · CheXpert->MIMIC-CXRAUC: +1.4-1.8 pointsImproved
add CS-TTA regularizerfive TTA baselines · CIFAR-10-C and ImageNet-Caccuracy: +0.6-1.4 pointsImproved
adapt with Tent, EATA or CoTTA (no concept regularization)source ResNet-50 · Waterbirdsworst-group accuracy: 58.1 -> 55.1 (CoTTA)Hurt
replace concept-selective loss with blind L2 on all activations (ablation)Tent+CS-TTA · Waterbirdsworst-group accuracy: 65.8 -> 57.5Improved
select random concepts instead of object/background concepts (ablation)Tent+CS-TTA · Waterbirdsworst-group accuracy: 65.8 -> 58.2Improved
raise lambda from 0.1 to 10.0 (ablation)Tent+CS-TTA · Waterbirdsworst-group accuracy: 65.8 -> 63.5Hurt
lower lambda from 0.1 to 0.01 (ablation)Tent+CS-TTA · Waterbirdsworst-group accuracy: 65.8 -> 62.3Hurt
regularize at early layer2 instead of layer4 (ablation)Tent+CS-TTA · Waterbirdsworst-group accuracy: 65.8 -> 62.8Hurt
use CLIP pseudo-label concepts instead of annotations (ablation)Tent+CS-TTA · Waterbirdsworst-group accuracy: 65.8 -> 64.5Hurt
use second-order exact sensitivity instead of first-order surrogate (ablation)Tent+CS-TTA · Waterbirdsworst-group accuracy: 65.8 -> 66.0No clear effect

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