Out-of-Distribution Detection on RACE (In-distribution) to MMLU (Out-of-distribution)
96.8MP ScoreETN
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| ETN2026.04 | 96.8 | 94.51 | 94.65 | |
| ETNBackbone=Gemma-2-9B2026.04 | 96.7 | 95.21 | 95.29 | |
| MCD2026.04 | 96.59 | 96.33 | — | |
| MCDBackbone=Gemma-2-9B2026.04 | 96.36 | 95.36 | — | |
| MAPCEBackbone=Gemma-2-9B2026.04 | 96.03 | — | — | |
| MAP_CE2026.04 | 96.01 | — | — | |
| DeepEnsBackbone=Gemma-2-9B2026.04 | 94.64 | 93.79 | — | |
| IB-EDL2026.04 | 94.45 | 85.95 | 94.38 | |
| DeepEns2026.04 | 94.28 | 91.42 | — | |
| MAPEDLBackbone=Gemma-2-9B2026.04 | 93.97 | 83.89 | 90.36 | |
| DMMBackbone=Gemma-2-9B2026.04 | 93.84 | 92.77 | 93.8 | |
| DMM2026.04 | 93.11 | 92.06 | 93.25 | |
| MAP_EDL2026.04 | 91.19 | 87.67 | 91.18 | |
| IB-EDLBackbone=Gemma-2-9B2026.04 | 90.67 | 89.43 | 90.7 | |
| LLBackbone=Gemma-2-9B2026.04 | 27.47 | 27.03 | — | |
| LL2026.04 | 25.24 | 27.22 | — |