Table of Contents of Appendix A Detailed Related Work 17 B Limitations and Potential Negative Societal Impacts 18 C Discussions and Details about Experiments in Figure 1 18 C.1 Summary

Distance-based methods are based on the assumption that OOD data should be relatively far away from the centroids of ID classes [9], including Mahalanobis distance [9, 45], cosine similarity [74], and kernel similarity [75]. Early works consider using the maximum softmax score to express the ID-ness [7]. Then, temperature scaling functions are used to amplify the separation between the ID and OOD data [14]. Recently, researchers propose hyperparameter-free energy scores to improve the OOD uncertainty estimation [23, 71]. Additionally, researchers also consider using the information contained in gradients to help improve the performance of OOD detection [18].