AI Paper Atlas

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

TANGO: Learning Distribution-wise Foundation Prior Consistency and Instance-wise Style Calibration for Medical Image Generalization

Chuang Liu, Yichao Cao, Xiu Su, Haogang Zhu

CVPR 2026 · Label-efficient and robust learning › Test-time adaptation · approach: Test-time adaptation of vision-language models

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 FAFPI foundation low-frequency prior injection (ablation)source model · Polyp segmentation (CTTA)Average DSC: 75.77 -> 78.34Improved
add foundation-informed consistency loss (FCL) (ablation)FAFPI source model · Polyp segmentation (CTTA)Average DSC: 78.34 -> 80.91Improved
add instance-wise style calibration at test time (ablation)DSCL-trained source model · Polyp segmentation (CTTA)Average DSC: 80.91 -> 83.70Improved
DSCL vs AttnKD feature distillation (ablation)source model training · polypAverage DSC: 79.02 -> 80.91Improved
DSCL vs AttnKD feature distillation (ablation)source model training · fundusAverage DSC: 81.32 -> 83.05Improved
add all three SKAR constraints (Lsas, Lsad, Lent) (ablation)LID test-time adaptation · Polyp segmentation (CTTA)Average DSC: 75.77 -> 83.70Improved
set low-frequency ratio r to 0.4 (vs 0 or higher) (ablation)FAFPI · polypDSCImproved
apply TENT continual adaptationPraNet source model · Polyp segmentation (CTTA)Average DSC: 75.77 -> 58.43Hurt
TanGo vs best baselinetest-time adaptation · Fundus OD/OC CTTAAverage DSC: 74.32 -> 85.23Improved

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