OOD Detection on WMT15 Subtitles OOD En-Fr ID (test)
95.41AUROCAP-OOD
Evaluation results
| Method | Links | ||
|---|---|---|---|
| AP-OODOOD level=Input, Learning paradigm=Supervised, Backbone=Transformer (base)2026.02 | 95.41 | 29.25 | |
| AP-OODOOD level=Output, Learning paradigm=Supervised, Backbone=Transformer (base)2026.02 | 95.27 | 30.02 | |
| Deep SADOOD level=Input, Learning paradigm=Supervised, Backbone=Transformer (base)2026.02 | 93.85 | 39 | |
| Binary logitsOOD level=Output, Learning paradigm=Supervised, Backbone=Transformer (base)2026.02 | 92.45 | 37.4 | |
| Relative MahalanobisOOD level=Output, Learning paradigm=Supervised, Backbone=Transformer (base)2026.02 | 92.42 | 30.83 | |
| Binary logitsOOD level=Input, Learning paradigm=Supervised, Backbone=Transformer (base)2026.02 | 91.56 | 41.74 | |
| AP-OODBackbone=Transformer (base), Training Data=WMT15 En-Fr, Evaluation Side=Output, Unsupervised=true2026.02 | 91.39 | 49.36 | |
| PerplexityBackbone=Transformer (base), Training Data=WMT15 En-Fr, Evaluation Side=Output, Unsupervised=true2026.02 | 91.27 | 51.31 | |
| Relative MahalanobisOOD level=Input, Learning paradigm=Supervised, Backbone=Transformer (base)2026.02 | 90.98 | 26.71 | |
| KNNBackbone=Transformer (base), Training Data=WMT15 En-Fr, Evaluation Side=Input, Unsupervised=true2026.02 | 87.47 | 71.91 | |
| AP-OODBackbone=Transformer (base), Training Data=WMT15 En-Fr, Evaluation Side=Input, Unsupervised=true2026.02 | 87.39 | 66.35 | |
| MahalanobisBackbone=Transformer (base), Training Data=WMT15 En-Fr, Evaluation Side=Input, Unsupervised=true2026.02 | 87.27 | 70.73 | |
| MahalanobisBackbone=Transformer (base), Training Data=WMT15 En-Fr, Evaluation Side=Output, Unsupervised=true2026.02 | 86.68 | 74.97 | |
| KNNBackbone=Transformer (base), Training Data=WMT15 En-Fr, Evaluation Side=Output, Unsupervised=true2026.02 | 85.87 | 79.04 | |
| Deep SVDDBackbone=Transformer (base), Training Data=WMT15 En-Fr, Evaluation Side=Input, Unsupervised=true2026.02 | 85.06 | 74.64 | |
| Deep SADOOD level=Output, Learning paradigm=Supervised, Backbone=Transformer (base)2026.02 | 84.72 | 80.49 | |
| Deep SVDDBackbone=Transformer (base), Training Data=WMT15 En-Fr, Evaluation Side=Output, Unsupervised=true2026.02 | 82.52 | 82.77 | |
| EntropyBackbone=Transformer (base), Training Data=WMT15 En-Fr, Evaluation Side=Output, Unsupervised=true2026.02 | 70.36 | 75.84 |