研究人员开发了新的方法来提高机器学习模型在测试时适应（TTA）的鲁棒性，尤其是在对抗性攻击和不断变化的数据分布场景下。一种名为SAFER的方法，利用随机增强和可靠性引导池来增强弹性，而无需源数据。另一个框架DO-ALL采用数据集蒸馏来创建用于稳定长期适应的合成锚点，通过避免保留原始源数据来解决隐私问题。此外，还提出了一个基于状态空间模型的概率框架用于在线TTA，以表征参数学习和演化。最后，双分布估计（DDE）提供了一种无需训练的方法，用于处理带有视觉语言模型的嘈杂TTA，提高了分布内准确性和分布外检测能力。
AI
arXiv:2607.00259v1 Announce Type: cross Abstract: Test-Time Adaptation (TTA) seeks to improve model robustness under distribution shifts by adapting parameters using unlabeled target data. However, in the absence of supervision, entropy-based adaptation is fundamentally undercons…
arXiv cs.AI
TIER_1English(EN)·Shaoyang Huang, Yashi Zhu, Yichen Yu, Lei Zhang, Zhang Yi, Tao He·
arXiv:2606.31420v1 Announce Type: new Abstract: Test-Time Adaptation (TTA) enables models trained on a source domain to adapt online to unlabeled test data under distribution shifts. While recent TTA methods have moved beyond static settings and begun to consider continual domain…
arXiv cs.LG
TIER_1English(EN)·Mansoo Jung, Youngwook Kim, Jungwoo Lee·
arXiv:2606.28142v1 Announce Type: new Abstract: Test-Time Adaptation (TTA) methods commonly update the affine parameters of normalization layers to adapt deployed models under distribution shifts. However, per-channel affine parameters perform axis-aligned scaling and shifting, m…
Test-Time Adaptation (TTA) methods commonly update the affine parameters of normalization layers to adapt deployed models under distribution shifts. However, per-channel affine parameters perform axis-aligned scaling and shifting, making them geometrically incapable of correcting…
This paper presents a probabilistic framework for online test-time adaptation problems. In them, a model is trained on labeled data but must adapt to unlabeled data at test time under the assumption that training and test distributions potentially differ, that is, there might hav…
Test-time adaptation (TTA) can mitigate domain shift without source data, but it is highly brittle under adversarially contaminated test streams, where corrupted inputs also destabilize online updates. We study robust test-time adaptation (RTTA) in the adversarial-stream setting,…
Test-time adaptation (TTA) can mitigate domain shift without source data, but it is highly brittle under adversarially contaminated test streams, where corrupted inputs also destabilize online updates. We study robust test-time adaptation (RTTA) in the adversarial-stream setting,…
DO-ALL is a test-time adaptation framework that uses dataset distillation to create synthetic anchors for stable long-term model performance without retaining source data.
arXiv stat.ML
TIER_1English(EN)·Daniel Corrales, David R\'ios Insua·
arXiv:2606.26457v1 Announce Type: new Abstract: This paper presents a probabilistic framework for online test-time adaptation problems. In them, a model is trained on labeled data but must adapt to unlabeled data at test time under the assumption that training and test distributi…
arXiv:2606.25758v1 Announce Type: new Abstract: While test-time adaptation (TTA) empowers vision-language models to adapt without costly retraining, it remains highly vulnerable to out-of-distribution (OOD) outliers prevalent in real-world applications. This discrepancy motivates…
This paper presents a probabilistic framework for online test-time adaptation problems. In them, a model is trained on labeled data but must adapt to unlabeled data at test time under the assumption that training and test distributions potentially differ, that is, there might hav…
While test-time adaptation (TTA) empowers vision-language models to adapt without costly retraining, it remains highly vulnerable to out-of-distribution (OOD) outliers prevalent in real-world applications. This discrepancy motivates Noisy TTA (NTTA), an online task to filter nois…
