TALOR：面向开放集测试时适应的即插即用特征校正模块
Tail Cues, Principal Corrections: Plug-and-Play Rectification for Open-Set Test-Time Adaptation
一篇方法论文：TALOR 不更新参数也不存记忆库，在开放集测试时适应上比 UniEnt 高 2.2% H-score，还能当插件给 COME 提分。
TALOR 是一个免反向传播的轻量校正模块，利用源模型线性头作为几何锚点，将归一化特征分解为高增益主坐标和低增益尾坐标，并通过加权岭回归估计偏移后进行修正。在 ImageNet-C 与六个 csOOD 数据集上，TALOR 的 H-score 比次优方法 UniEnt 高出 2.2%，同时速度快 2.5 倍、GPU 显存仅为其 12%。作为后处理插件接入 COME 时，可将 H-score 提升至 68.4%，额外显存开销仅 2%。
Tail Cues, Principal Corrections: Plug-and-Play Rectification for Open-Set Test-Time Adaptation
Open-Set Test-Time Adaptation (OSTTA) operates on non-stationary target streams where covariate and semantic shifts coexist. Existing TTA methods often obtain adaptation signals by updating model parameters or maintaining mutable memories, which can be costly and vulnerable to unknown samples. We present TALOR, a lightweight, backward-free rectification module that uses the source linear head as a geometric anchor. The head-induced basis decomposes each normalized feature into high-gain principal coordinates and low-gain tail coordinates. TALOR estimates a soft-routed affine correction in the principal subspace, combining a regime-level principal bias with a tail-conditioned slope that captures sample-level tail-to-principal coupling. Using weighted ridge regression, it subtracts this drift estimate from the principal coordinates, reconstructs the feature and feeds it into the source head. On the ImageNet-C benchmark with six csOOD datasets, TALOR outperforms the next-best UniEnt by 2.2% H-score while running 2.5x faster and using only 12% of its GPU memory; as a post-correction plug-in, it further boosts COME to 68.4% H-score with only 2% additional memory overhead.
