AI Paper Atlas

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

ClusT3: Information Invariant Test-Time Training

Gustavo A. Vargas Hakim, David Osowiechi, Mehrdad Noori, Milad Cheraghalikhani, Ali Bahri, Ismail Ben Ayed, Christian Desrosiers

ICCV 2023 · Label-efficient and robust learning › Test-time adaptation · approach: Test-time adaptation of vision-language models · 18 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 multi-head clustering (15 projectors per layer vs 1) (ablation)ClusT3 on layers 1-2 · CIFAR-10-C Level 5 (average)accuracy (%): 77.97 -> 82.08Improved
increase heads from 15 to 20 per layer (ablation)ClusT3 on layers 1-2 · CIFAR-10-C Level 5 (average)accuracy (%): 82.08 -> 81.73Hurt
place projectors on early layers (1-2) rather than later layers (3-4) (ablation)ClusT3 · CIFAR-10-C Level 5 (Gaussian Noise)accuracy (%): 65.21 -> 71.36Improved
add projectors on all four layers instead of only layers 1-2 (ablation)ClusT3 · CIFAR-10-C Level 5 (Gaussian Noise)accuracy (%): 71.36 -> 68.71Hurt
use K=10 clusters rather than K=100 (ablation)ClusT3 · CIFAR-10-C Level 5 (average)accuracy (%): 79.77 -> 80.94Improved
vary number of clusters K from 2 to 50 (ablation)ClusT3 · CIFAR-10-C Level 5 (average)accuracy (%): 80.39 (K=2) vs 80.94 (K=10) vs 80.70 (K=50)No clear effect
apply ClusT3 test-time adaptation vs no adaptationResNet50 · CIFAR-10-C Level 5 (average)accuracy (%): 53.82 -> 82.08Improved
adapt with ClusT3 under small natural shiftResNet50 · CIFAR-10.1accuracy (%): 88.45 -> 85.57Hurt
adapt with ClusT3 (5 heads) under small natural shiftResNet50 · CIFAR-10.1accuracy (%): 88.45 -> 87.43Hurt
apply ClusT3 test-time adaptation on sim-to-real shiftResNet50 · VisDA-Caccuracy (%): 46.31 -> 61.91Improved
use ClusT3 instead of TTT auxiliary task for joint trainingTTT-style joint training · CIFAR-10 joint training time per epochminutes per epoch: 12.4941 -> 2.7947Improved
ClusT3-H15 vs TTT per-corruptionCIFAR-10-C Level 5 · CIFAR-10-C Level 5 (Contrast)accuracy (%): 88.58 (TTT) vs 84.40 (ClusT3-H15)Mixed

Later papers that compared with this method

2 later accepted papers report ClusT3 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