AI Paper Atlas

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

In Search of Lost Online Test-Time Adaptation: A Survey

Zixin Wang, Yadan Luo, Liang Zheng, Zhuoxiao Chen, Sen Wang, Zi Helen Huang

IJCV 2024 · Label-efficient and robust learning › Test-time adaptation · approach: Realistic and stable test-time adaptation · 54 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
update all parameters instead of only LayerNorm (ablation)CoTTA · CIFAR-10-C severity 5 (batch size 16)error rate (%): 24.71 -> 52.47Hurt
add parameter reset to full-network updating (ablation)CoTTA-ALL · CIFAR-10-C severity 5 (batch size 16)error rate (%): 77.08 -> 52.47Improved
add parameter reset to LayerNorm-only updating (ablation)CoTTA-LN · CIFAR-10-C severity 5 (batch size 16)error rate (%): 24.68 -> 24.71No clear effect
add parameter reset to full-network updating (ablation)CoTTA-ALL · ImageNet-C severity 5 (batch size 16)error rate (%): 90.54 -> 85.16Improved
use LayerNorm-only updating instead of full update (ablation)CoTTA · ImageNet-C severity 5 (batch size 16)error rate (%): 85.16 -> 72.37Improved
apply Tent entropy minimization to ViTTent · CIFAR-10-C severity 5 (batch size 16)error rate (%): 23.89 -> 24.90Hurt
apply Tent entropy minimization to ViTTent · ImageNet-C severity 5 (batch size 16)error rate (%): 63.24 -> 77.66Hurt
apply ROID to ViTROID · CIFAR-10-C severity 5 (batch size 16)error rate (%): 23.89 -> 13.36Improved
apply ROID to ViTROID · ImageNet-C severity 5 (batch size 16)error rate (%): 63.24 -> 46.70Improved
use batch size 1 instead of 16 for ROID (ablation)ROID · CIFAR-10-C severity 5error rate (%): 13.36 -> 89.99Hurt
use batch size 1 instead of 16 for Tent (ablation)Tent · CIFAR-10-C severity 5error rate (%): 24.90 -> 52.25Hurt
apply SAR (sharpness-aware) on top of Tent-style updateSAR · ImageNet-C severity 5 (batch size 16)error rate (%): 77.66 -> 52.70Improved

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