AI Paper Atlas

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

Architecture-Agnostic Test-Time Adaptation via Backprop-Free Embedding Alignment

MA Xiao, Young D. Kwon, Pan Zhou, Dong Ma

ICLR 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 layer-wise distance-based weighting to covariance alignment (ablation)PEA ViT-Base · ImageNet-Caccuracy (%): 25.2 -> 52.9Improved
add layer-wise weighting (ablation)PEA ViT-Base · CIFAR100-Caccuracy (%): 67.0 -> 68.3Improved
add EMA of target statistics (ablation)PEA ViT-Base · CIFAR100-Caccuracy (%): 68.3 -> 75.7Improved
add EMA of target statistics (ablation)PEA ViT-Base · ImageNet-Caccuracy (%): 52.9 -> 64.5Improved
add lightweight augmentation views (ablation)PEA ViT-Base · ImageNet-Caccuracy (%): 64.5 -> 66.5Improved
apply covariance alignment to all layers without weighting (ablation)PEA ViT-Base · ImageNet-Caccuracy (%): 55.5 -> 25.2Hurt
apply covariance alignment only (no weighting) (ablation)PEA ViT-Base · CIFAR100-Caccuracy (%): 61.6 -> 67.0Improved
use entropy minimization (Tent, EATA) on mixed-domain batchesTent/EATA ViT-Base · CIFAR100-C mixed domainaccuracy (%): 61.6 -> 61.2No clear effect
use entropy minimization (Tent) on mixed-domain batchesTent ResNet-50 · CIFAR100-C mixed domainaccuracy (%): 33.5 -> 17.4Hurt
reduce batch size from 64 to 4PEA ViT-Base · CIFAR100-Caccuracy (%): 77.0 -> 70.0Hurt
use backprop-free PEA instead of backprop-based TentPEA ViT-Base · ImageNet-C memorymemory (MB): 6108 -> 887Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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