通过熵敏感度引导，在单批次在线场景下实现稳定持续的测试时自适应。
Continual Test-Time Adaptation via Entropy Sensitivity-Guidance in Strict Online Setting
- 用结构化擦除探测预测熵变化，生成样本敏感度轨迹。
- 在多个图像数据集上显著提升鲁棒性与稳定性，优于现有方法。
- 无需周期重置或模型缓存，适合长期单批次在线学习场景。
测试时自适应（TTA）通过在无标签测试数据上更新预训练模型来提升分布偏移下的鲁棒性，但在仅单批次且无源数据访问的严格在线设置下极易发生性能漂移或崩溃。本文提出敏感度引导擦除自适应（SEGA），用于在腐蚀型数据流上的严格在线持续测试时自适应（CTTA）。SEGA利用少量结构化擦除，探测信息移除时预测熵的变化，基于每样本的敏感度轨迹协调恢复与样本选择，而非依赖原始熵或批量统计。该方法在不需周期重置或模型缓存的前提下，为长时程单批次适应提供有效反馈信号。在ImageNet-C、CIFAR10/100-C及模拟水产养殖腐蚀流上实验表明，SEGA相比强基线持续提升鲁棒性与稳定性，同时通过敏感度门控减少反向传播次数。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。