Authors
Shuaicheng Niu, Guohao Chen, Deyu Chen, Yifan Zhang, Jiaxiang Wu, Zhiquan Wen, Yaofo Chen, Peilin Zhao, Chunyan Miao, Mingkui Tan
Published in
IEEE transactions on pattern analysis and machine intelligence. Volume PP. Aug 11, 2026. Epub Aug 11, 2026.
Abstract
Test-time adaptation (TTA) has shown to be effective at tackling distribution shifts between training and testing data by adapting a given model on test samples. However, the online model updating of TTA may be unstable and this is often a key obstacle preventing existing TTA methods from being deployed in the real world. Specifically, TTA may fail to improve or even harm the model performance when test data have: 1) mixed distribution shifts, 2) small batch sizes, and 3) online imbalanced label distribution shifts, which are quite common in practice. In this paper, we investigate the unstable reasons and find that the batch norm layer is a crucial factor hindering TTA stability. Conversely, TTA can perform more stably with batch-agnostic norm layers, i.e., group or layer norm. However, we observe that TTA with group and layer norms does not always succeed and still suffers many failure cases, i.e., the model collapses into trivial solutions by assigning the same class label for all samples. By digging into this, we find that, during the collapse process: 1) the model gradients often undergo an initial explosion followed by rapid degradation, suggesting that certain noisy test samples with large gradients may disrupt adaptation; and 2) the model representations tend to exhibit high correlations and classification bias. To address the above collapse issue, we first propose a sharpness-aware and reliable entropy minimization method, called SAR, for stabilizing TTA from two aspects: 1) remove partial noisy samples with large gradients, 2) encourage model weights to go to a flat minimum so that the model is robust to the remaining noisy samples. Based on SAR, we further introduce SAR2 to prevent representation collapse with two regularizers: 1) a redundancy regularizer, which reduces inter-dimensional correlations among centroid-invariant features; and 2) an inequity regularizer, which maximizes the prediction entropy of a prototype centroid, thereby penalizing biased representations toward any specific class. Promising results demonstrate that our methods perform more stably over prior methods and are computationally efficient under the above wild test scenarios. The source code is available at https://github.com/mr-eggplant/SAR.
PMID:
42579581
Bibliographic data and abstract were imported from PubMed on 12 Aug 2026.
Read full publication at:
Please sign in
to see all details.
Advertisement
Stats
- Recommendations n/a n/a positive of 0 vote(s)
- Views 3
- Comments 0