Label Shift Adapter for Test-Time Adaptation under Covariate and Label Shifts
Sunghyun Park, Seunghan Yang, Jaegul Choo, Sungrack Yun

TL;DR
This paper introduces a label shift adapter for test-time adaptation that effectively handles both covariate and label distribution shifts, improving model performance in real-world domain changes.
Contribution
The proposed label shift adapter is a novel, computationally efficient method that estimates and incorporates target label distributions into existing TTA approaches.
Findings
Significant performance improvements under joint covariate and label shifts.
Effective estimation of target label distribution during inference.
Easy integration with various model architectures.
Abstract
Test-time adaptation (TTA) aims to adapt a pre-trained model to the target domain in a batch-by-batch manner during inference. While label distributions often exhibit imbalances in real-world scenarios, most previous TTA approaches typically assume that both source and target domain datasets have balanced label distribution. Due to the fact that certain classes appear more frequently in certain domains (e.g., buildings in cities, trees in forests), it is natural that the label distribution shifts as the domain changes. However, we discover that the majority of existing TTA methods fail to address the coexistence of covariate and label shifts. To tackle this challenge, we propose a novel label shift adapter that can be incorporated into existing TTA approaches to deal with label shifts during the TTA process effectively. Specifically, we estimate the label distribution of the target…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsDomain Adaptation and Few-Shot Learning · Cancer-related molecular mechanisms research
Methodsfail · Adapter
