Rank-1 权重修正的可训练权重平均方法用于 OOD 检测
Rank-1 Weight-Corrected Trainable Weight Averaging for Out-of-Distribution Detection
一篇把权重平均和秩分析结合改进 OOD 检测的论文，ResNet50 上跑出了对比数据，做模型部署异常检测的可以看看。
论文基于 RankFeat 和 RankWeight 的秩分析，认为 OOD 过度自信与特征和权重空间中的主导低秩分量相关。作者在 Trainable Weight Averaging（TWA）基础上，用 SVD 提取的 leading rank-one 权重分量作为修正方向，其贡献通过分布内分类目标学习得到。在 ResNet50 上，该方法在使用 MSP 和 RankWeight 评分时有和无 Outlier Exposure 的数据集上都优于 fine-tuning 和标准 TWA。Mahalanobis 评测结果不一，说明收益取决于 OOD 评分协议；MSP 和 Energy 的目标层分析显示修正靠近分类器层更有效。
Rank-1 Weight-Corrected Trainable Weight Averaging for Out-of-Distribution Detection
Out-of-distribution (OOD) detection identifies inputs outside the training distribution. Rank-based analyses such as RankFeat and RankWeight suggest that OOD overconfidence is closely related to dominant low-rank components in feature and weight spaces. Since weight averaging operates directly in parameter space and improves model generalization, we consider it for OOD detection from a dominant-component perspective. Trainable Weight Averaging (TWA) assigns a single coefficient to each checkpoint, limiting independent control of dominant weight components. We extend TWA with leading rank-one weight components extracted by Singular Value Decomposition as correction directions, whose contributions are learned through the in-distribution classification objective. On ResNet50, our method improves over fine-tuning and standard TWA across the evaluated datasets with MSP and RankWeight, both with and without Outlier Exposure. Mahalanobis-based evaluation yields mixed results, indicating that the benefit depends on the OOD scoring protocol. Target-layer analyses with MSP and Energy favor correction near the classifier.