Research map / Label-efficient and robust learning / OOD and anomaly detection
Characterizing Out-of-Distribution Error via Optimal Transport
Yuzhe Lu, Yilong Qin, Runtian Zhai, Andrew Shen, Ketong Chen, Zhenlin Wang, Soheil Kolouri, Simon Stepputtis, Joseph Campbell, Katia P. Sycara
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| COT (confidence plus pseudo-label shift via optimal transport) instead of average confidence | AC · ImageNet Natural | MAE: 8.5 -> 3.88 | Improved |
| COT instead of average confidence | AC · CIFAR100 Synthetic | MAE: 10.83 -> 2.09 | Improved |
| COT instead of average confidence | AC · Nonliving26 Novel | MAE: 31.37 -> 17.8 | Improved |
| COT instead of average confidence | AC · CIFAR10 Natural | MAE: 5.97 -> 5.41 | Improved |
| COTT (thresholded transport cost) instead of COT | COT · Entity13 Novel | MAE: 15.9 -> 5.46 | Improved |
| COTT instead of ATC-NE on synthetic shift | ATC-NE · CIFAR10 Synthetic | MAE: 3.37 -> 1.7 | Improved |
| COTT instead of ATC-NE on novel subpopulation shift | ATC-NE · Nonliving26 Novel | MAE: 16.56 -> 10.46 | Improved |
| COTT instead of ATC-NE on natural shift | ATC-NE · CIFAR10 Natural | MAE: 3.15 -> 3.33 | Hurt |
| COTT instead of ATC-NE on ImageNet natural shift | ATC-NE · ImageNet Natural | MAE: 2.6 -> 2.41 | Improved |
| COTT instead of ATC-NE on label-shifted CivilComments | ATC-NE · CivilComments-WILDS | MAE: 2.3 -> 4.68 | Hurt |
| COTT instead of ATC-NE on Camelyon17 | ATC-NE · Camelyon17-WILDS | MAE: 7.73 -> 5.71 | Improved |
| COT instead of ATC-NE on RxRx1 | ATC-NE · RxRx1-WILDS | MAE: 6.86 -> 3.25 | Improved |
Closest papers by meaning
- Bypassing the Transport Plan: Dynamic Reweighting for Out-of-Distribution Detection with Optimal Transport (CVPR 2026 · similarity 0.88)
- Optimal Transport Reweighting for Robust Learning under Spurious Correlations and Label Noise (NeurIPS 2026 · similarity 0.86)
- Estimation of Stochastic Optimal Transport Maps (NeurIPS 2025 · similarity 0.85)
- Expecting The Unexpected: Towards Broad Out-Of-Distribution Detection (NeurIPS 2024 · similarity 0.85)
- Stability Evaluation through Distributional Perturbation Analysis (ICML 2024 · similarity 0.85)
- Aggregation Hides Out-of-Distribution Generalization Failures from Spurious Correlations (NeurIPS 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
