Machine Reading Comprehension on Belebele Source language en
89.8MRC ScoreAdaLoRA
Evaluation results
| Method | Links | |
|---|---|---|
| AdaLoRAModel Scale=13B, Evaluation Protocol=3-shot2025.12 | 89.8 | |
| SSU-RandModel Scale=13B, Evaluation Protocol=3-shot2025.12 | 89.8 | |
| SourceModel Scale=13B, Evaluation Protocol=3-shot2025.12 | 89.7 | |
| SSU-WandaModel Scale=13B, Evaluation Protocol=3-shot2025.12 | 89.7 | |
| GMTModel Scale=13B, Evaluation Protocol=3-shot2025.12 | 89.6 | |
| HFTModel Scale=13B, Evaluation Protocol=3-shot2025.12 | 89.4 | |
| FFTModel Scale=13B, Evaluation Protocol=3-shot2025.12 | 89.2 | |
| SSU-MagModel Scale=13B, Evaluation Protocol=3-shot2025.12 | 89.2 | |
| SourceModel Scale=7B, Evaluation Protocol=3-shot2025.12 | 88 | |
| SSU-WandaModel Scale=7B, Evaluation Protocol=3-shot2025.12 | 86.9 | |
| AdaLoRAModel Scale=7B, Evaluation Protocol=3-shot2025.12 | 86.6 | |
| SSU-RandModel Scale=7B, Evaluation Protocol=3-shot2025.12 | 86.1 | |
| SSU-MagModel Scale=7B, Evaluation Protocol=3-shot2025.12 | 85.9 | |
| HFTModel Scale=7B, Evaluation Protocol=3-shot2025.12 | 85.4 | |
| FFTModel Scale=7B, Evaluation Protocol=3-shot2025.12 | 84.2 | |
| GMTModel Scale=7B, Evaluation Protocol=3-shot2025.12 | 84.1 |