ServiceNow Research

Out of Distribution

Expecting The Unexpected: Towards Broad Out-Of-Distribution Detection
Deployed machine learning systems require some mechanism to detect out-of-distribution (OOD) inputs. Existing research mainly focuses …
CADet: Fully Self-Supervised Out-Of-Distribution Detection With Contrastive Learning
Handling out-of-distribution (OOD) samples has become a major stake in the real-world deploy- ment of machine learning systems. This …
Group Robust Classification Without Any Group Information
Empirical risk minimization (ERM) is sensitive to spurious correlations present in training data, which poses a significant risk when …