Natural Language Inference on AmericasNLI (dev)
48.8Accuracy (Aymara)MindMerger
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| MindMerger2025.09 | 48.8 | — | 48.7 | 48.3 | 48.6 | |
| MERLIN2025.09 | 48.4 | — | 49.3 | 50.5 | 49.4 | |
| LLaMAX2 XNLI (base)2025.09 | 41.6 | — | 43.6 | 40.7 | 42 |
| Method | Links | |||||
|---|---|---|---|---|---|---|
| MindMerger2025.09 | 48.8 | — | 48.7 | 48.3 | 48.6 | |
| MERLIN2025.09 | 48.4 | — | 49.3 | 50.5 | 49.4 | |
| LLaMAX2 XNLI (base)2025.09 | 41.6 | — | 43.6 | 40.7 | 42 |