AI Paper Atlas

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

Improved Test-Time Adaptation for Domain Generalization

Liang Chen, Yong Zhang, Yibing Song, Ying Shan, Lingqiao Liu

CVPR 2023 · Label-efficient and robust learning › Test-time adaptation · approach: Realistic and stable test-time adaptation · 44 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 a learnable weight subnetwork to the consistency loss (ablation)ITTA · PACSaccuracy: 82.5±0.1 -> 83.8±0.3Improved
replace learnable consistency with entropy minimization (ablation)ITTA · PACSaccuracy: 83.8±0.3 -> 82.4±0.8Hurt
replace learnable consistency with rotation estimation (ablation)ITTA · PACSaccuracy: 83.8±0.3 -> 82.1±0.3Hurt
add test-time updating of the new adaptive parameters (ablation)ITTA · PACSaccuracy: 82.8±0.3 -> 83.8±0.3Improved
update all original feature-extractor parameters instead of the added adaptive parameters (ablation)ITTA · PACSaccuracy: 83.8±0.3 -> 83.0±0.2Hurt
update only batch-normalization parameters instead of the added adaptive parameters (ablation)ITTA · PACSaccuracy: 83.8±0.3 -> 82.4±0.5Hurt
increase the number of TTT updates from one to three (ablation)ITTA · PACSaccuracy: 83.8 ± 0.3 -> 83.9 ± 0.3No clear effect
switch from online to episodic updating (ablation)ITTA · PACSaccuracy: 83.8 ± 0.3 -> 83.3 ± 0.4Hurt
vary the depth of the weight subnetwork or adaptive blocks (ablation)ITTA · PACSaccuracy: 83.1 ± 0.6 to 83.8 ± 0.3No clear effect
use a learnable consistency loss rather than naive consistency with the alternative augmentation (ablation)ITTA · PACSaccuracy: 79.7 ± 0.2 -> 80.8 ± 0.4Improved

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