Multi-task Language Understanding on GMMLU Spanish c
33.75Acc (Normalized)HQ seed
Evaluation Results
| Method | Links | |
|---|---|---|
| HQ seedFiltering strategy=HQ seed2026.04 | 33.75 | |
| ML (Q3)Filtering strategy=ML (Q3)2026.04 | 32.75 | |
| MLFiltering strategy=ML2026.04 | 32.5 | |
| ML (15%)Filtering strategy=ML (15%)2026.04 | 32.25 | |
| MKC-eFiltering strategy=MKC-e2026.04 | 32 | |
| HQFiltering strategy=HQ2026.04 | 31.5 | |
| No filteringFiltering strategy=No filtering2026.04 | 31 |