发表机构
Norwegian University of Life Sciences (NMBU); Tulane University（挪威生命科学大学; 杜兰大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出SEGA方法，用于严格在线持续测试时适应，通过结构化擦除探测熵变化以协调恢复与样本选择，在多数据集上提升了鲁棒性与稳定性并减少反向传播次数。
AI 中文摘要
测试时适应（TTA）通过在无标签测试数据上更新预训练模型，可实现分布偏移下的鲁棒性，但批量大小为1且无法访问源数据的严格在线TTA尤其容易出现漂移或崩溃。本文提出Sensitivity-Guided Erasing Adaptation（SEGA，敏感性引导擦除适应），一种针对损坏式数据流的严格在线持续TTA（CTTA）方法。SEGA使用少量结构化擦除操作探测移除信息时预测熵的变化，并利用得到的逐样本敏感性轨迹协调恢复与样本选择，而非依赖原始熵或批量统计量，为无需周期性重置或模型储备的长周期批量大小为1的适应提供实用反馈信号。在ImageNet-C、CIFAR10/100-C及作为受控损坏式代理的损坏生成水产数据流上的实验显示，SEGA较强大的CTTA基线实现了一致的鲁棒性与稳定性提升，同时通过敏感性门控减少了反向传播次数。
英文摘要
Test-time adaptation (TTA) promises robustness under distribution shift by updating a pretrained model on unlabeled test data, but strict online TTA with batch size one and no access to source data is especially prone to drift or collapse. We introduce Sensitivity-Guided Erasing Adaptation (SEGA), a method for strict online continual TTA (CTTA) on corruption-style streams. SEGA uses a small number of structured erasures to probe how predictive entropy changes as information is removed, and uses the resulting per-sample sensitivity trajectories to coordinate recovery and sample selection rather than relying on raw entropy or batch statistics. This yields a practical feedback signal for long-horizon batch-size-one adaptation without periodic resets or model reservoirs. In experiments on ImageNet-C, CIFAR10/100-C, and corruption-generated aquaculture streams treated as controlled corruption-style proxies, SEGA yields consistent robustness and stability gains over strong CTTA baselines while reducing backward passes through sensitivity-based gating.
Commentsunder review. code available at https://github.com/chandlerbing65nm/SEGA.git
