AI Paper Atlas

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

CAFA: Class-Aware Feature Alignment for Test-Time Adaptation

Sanghun Jung, Jungsoo Lee, Nanhee Kim, Amirreza Shaban, Byron Boots, Jaegul Choo

ICCV 2023 · Label-efficient and robust learning › Test-time adaptation · approach: Pseudo-label test-time adaptation · 20 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
replace class-agnostic global feature alignment with class-aware (CAFA) alignment (ablation)TTA feature alignment · CIFAR10-Cclassification error (%): 19.12 -> 12.13Improved
add inter-class distance enlargement to intra-class alignment (ablation)CAFA · CIFAR10-Cclassification error (%): 13.02 -> 12.13Improved
add inter-class distance enlargement to intra-class alignment (ablation)CAFA · ImageNet-Cclassification error (%): 59.12 -> 55.24Improved
update only BN affine parameters instead of all feature extractor parameters (ablation)CAFA · CIFAR10-Cclassification error (%): 12.66 -> 12.13Improved
update all parameters vs BN only on ImageNet-C (ablation)CAFA · ImageNet-Cclassification error (%)No clear effect
use class-wise covariance instead of tied covariance (ablation)CAFA · CIFAR10-Cclassification error (%): 12.47 -> 12.13Improved
use Mahalanobis distance instead of Euclidean distance (ablation)CAFA · ImageNet-Cclassification error (%): 56.30 -> 55.24Improved
use Mahalanobis distance instead of class-aware KL-divergence (ablation)CAFA · CIFAR10-Cclassification error (%): 12.53 -> 12.13Improved
use CAFA instead of TENT under label shifttest-time adaptation · CIFAR10-C, label shift s=1.0classification error (%): 26.01 -> 17.61Improved
use CAFA instead of EATA on ImageNet-Ctest-time adaptation · ImageNet-Cclassification error (%): 52.00 -> 55.24Hurt
estimate source statistics from about 5% of training samples (ablation)CAFA · CIFAR10-C and ImageNet-Cclassification error (%)No clear effect
use CAFA instead of TENT at batch size 1test-time adaptation · CIFAR10-C, batch size 1classification error (%): 17.66 -> 15.00Improved

Later papers that compared with this method

2 later accepted papers report CAFA as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

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