Papers › Test-Time Training with Self-Supervision for Generalization under Distribution Shifts

Test-Time Training with Self-Supervision for Generalization under Distribution Shifts

29 Sep 2019arXiv:1909.13231archive 2025-07-28

Yu Sun, Xiaolong Wang, Zhuang Liu, John Miller, Alexei A. Efros, Moritz Hardt

In this paper, we propose Test-Time Training, a general approach for improving the performance of predictive models when training and test data come from different distributions. We turn a single unlabeled test sample into a self-supervised learning problem, on which we update the model parameters before making a prediction. This also extends naturally to data in an online stream. Our simple approach leads to improvements on diverse image classification benchmarks aimed at evaluating robustness to distribution shifts.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

yueatsprograms/ttt_cifar_release officialmentioned on GitHubpytorch report
yueatsprograms/ttt_imagenet_release officialmentioned on GitHubpytorch report
tejas-gokhale/AGAT mentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Building change detection for remote sensing imagesCARLA MAP LeaderboardImage ClassificationLanguage ModellingLow-Light Image EnhancementSelf-Supervised LearningSupervised Video SummarizationVideo Quality Assessmentimage-classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Language Modelling LAMBADA test Accuracy 0.01 #34 of 37 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections