AI Paper Atlas

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

SoTTA: Robust Test-Time Adaptation on Noisy Data Streams

Taesik Gong, Yewon Kim, Taeckyung Lee, Sorn Chottananurak, Sung-Ju Lee

NeurIPS 2023 · Label-efficient and robust learning › Test-time adaptation · approach: Pseudo-label test-time adaptation · 13 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 entropy-sharpness minimization (ESM) to HUS (ablation)SoTTA · CIFAR10-C avg over scenariosaccuracy (%): 57.9 -> 81.9Improved
add HUS to ESM (ablation)SoTTA · CIFAR10-C Noiseaccuracy (%): 60.5 -> 80.0Improved
add uniform-class sampling to ESM (ablation)SoTTA · CIFAR10-C avg over scenariosaccuracy (%): 75.4 -> 79.6Improved
add high-confidence sampling to UC + ESM (full HUS) (ablation)SoTTA · CIFAR10-C avg over scenariosaccuracy (%): 79.6 -> 81.9Improved
high-confidence sampling only, no class balancing or ESM (ablation)SoTTA · CIFAR10-C avg over scenariosaccuracy (%): 57.7 -> 20.0Hurt
add high-confidence sampling to uniform-class sampling without ESM (ablation)SoTTA · CIFAR10-C Attackaccuracy (%): 70.0 -> 40.9Hurt
replace entropy minimization with ESM, no sampling (ablation)SoTTA · CIFAR10-C Benignaccuracy (%): 69.8 -> 82.6Improved
SoTTA vs best baseline under NoiseSoTTA · CIFAR10-C Noiseaccuracy (%): 73.6 -> 80.0Improved
SoTTA vs best baseline without noisy samplesSoTTA · CIFAR10-C Benignaccuracy (%): 82.4 -> 82.2No clear effect
mix noisy samples into test streamTENT · CIFAR10-Caccuracy (%): 81.0 -> 52.1Hurt
mix noisy samples into test streamSoTTA · CIFAR10-Caccuracy (%): 82.2 -> 80.0Hurt
increase noisy samples from 5k to 20k (ablation)SoTTA · CIFAR10-C Noiseaccuracy (%): 1.9%p degradationNo clear effect

Later papers that compared with this method

5 later accepted papers report SoTTA as a baseline in their results tables; 5 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