Text Classification on MultiNLI (test)
81.3WGADPE
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| DPEGroup Info (Train/Val)=✗/✓, Group labels usage in training=used2026.02 | 81.3 | 74.8 | |
| DFRGroup Info (Train/Val)=✗/✓✓, Group labels usage in training=used2026.02 | 74.7 | 82.1 | |
| AFRGroup Info (Train/Val)=✗/★, Group labels usage in training=not used2026.02 | 73.4 | 81.4 | |
| AFRGroup Info (Train/Val)=✗/★2026.02 | 73.4 | 81.4 | |
| JTTGroup Info (Train/Val)=✗/★, Group labels usage in training=not used2026.02 | 72.6 | 78.6 | |
| JTTGroup Info (Train/Val)=✗/★2026.02 | 72.6 | 78.6 | |
| Group DROGroup Info (Train/Val)=✓/★, Group labels usage in training=used2026.02 | 69.9 | 88.9 | |
| LEIAGroup Info (Train/Val)=✗/★, Group labels usage in training=not used2026.02 | 69.6 | 81.4 | |
| LEIAGroup Info (Train/Val)=✗/★2026.02 | 69.6 | 81.4 | |
| ERM*Group Info (Train/Val)=✗/★, Group labels usage in training=not used2026.02 | 69.5 | 80.9 | |
| ERMGroup Info (Train/Val)=✗/★2026.02 | 69.5 | 80.9 | |
| CRT*Group Info (Train/Val)=✗/★, Group labels usage in training=not used2026.02 | 65.4 | 80.2 | |
| CRTGroup Info (Train/Val)=✗/★2026.02 | 65.4 | 80.2 | |
| ReWeightCRT*Group Info (Train/Val)=✗/★, Group labels usage in training=not used2026.02 | 65.3 | 80.2 | |
| ReWeightCRTGroup Info (Train/Val)=✗/★2026.02 | 65.3 | 80.2 | |
| IRM*Group Info (Train/Val)=✓/★, Group labels usage in training=used2026.02 | 63.6 | 77.8 | |
| CVaRDRO*Group Info (Train/Val)=✗/★, Group labels usage in training=not used2026.02 | 63 | 75.1 | |
| CVaRDROGroup Info (Train/Val)=✗/★2026.02 | 63 | 75.1 |