AI Paper Atlas

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

MixTTA: Low-Rank Cross-Channel Mixing for Reliable Test-Time Adaptation

Mansoo Jung, Youngwook Kim, Jungwoo Lee

ECCV 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 MixTTA low-rank cross-channel mixingTent (ResNet-GN) · ImageNet-C sev5 mildaccuracy: 28.8 -> 32.4Improved
add MixTTAEATA (ResNet-GN) · ImageNet-C sev5 mildaccuracy: 45.5 -> 46.2Improved
add MixTTASAR (ResNet-GN) · ImageNet-C sev5 mildaccuracy: 37.2 -> 37.5No clear effect
add MixTTADeYO (ResNet-GN) · ImageNet-C sev5 mildaccuracy: 45.3 -> 44.4Hurt
add MixTTAReCAP (ResNet-GN) · ImageNet-C sev5 mildaccuracy: 46.8 -> 47.9Improved
add MixTTA under online imbalanced label shiftsEATA (ResNet-GN) · ImageNet-C sev5 label shiftsaccuracy: 33.1 -> 41.3Improved
add MixTTA under label shiftsDeYO (ResNet-GN) · ImageNet-C sev5 label shiftsaccuracy: 42.3 -> 42.1No clear effect
add MixTTA at batch size 1Tent / EATA (ResNet-GN) · ImageNet-C sev5 batch size 1accuracy: Tent 23.9 -> 30.2; EATA 37.2 -> 41.5Improved
add MixTTA at batch size 1DeYO (ResNet-GN) · ImageNet-C sev5 batch size 1accuracy: 47.0 -> 45.4Hurt
add MixTTA under mixed corruption shiftsTent (ResNet-GN) · ImageNet-C sev5 mixed shiftsaccuracy: 14.1 -> 10.5Hurt
add MixTTA under mixed corruption shiftsDeYO (ResNet-GN) · ImageNet-C sev5 mixed shiftsaccuracy: 37.2 -> 40.1Improved
add MixTTA on stylized domain gap at batch size 1Tent / EATA / SAR (ResNet-GN) · ImageNet-Sketch batch size 1accuracy: Tent 28.5 -> 32.0; EATA 29.5 -> 33.3; SAR 31.1 -> 33.8Improved

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