Relevance Classification on ESCI-US (test)
70.23AccuracyMPCoT
Evaluation results
| Method | Links | ||
|---|---|---|---|
| MPCoTModel Category=Teacher Model, Protocol=SFT+DPO2026.01 | 70.23 | 51.85 | |
| Best-SingleCoTModel Category=Teacher Model, Protocol=SFT+DPO2026.01 | 69.49 | 50.08 | |
| LRKDModel Category=Student Model, Architecture=GAT2026.01 | 68.73 | 45.83 | |
| MKDModel Category=Student Model2026.01 | 68.34 | 42.45 | |
| LRKDModel Category=Student Model, Architecture=Poly2026.01 | 67.92 | 44.33 | |
| CED-KDModel Category=Student Model2026.01 | 67.49 | 43.71 | |
| LRKDModel Category=Student Model, Architecture=MLP2026.01 | 66.94 | 42.61 | |
| BERT-multilingual-baseModel Category=Deployable Model Baseline2026.01 | 66.83 | 41.38 | |
| MPCoTModel Category=Teacher Model, Protocol=SFT2026.01 | 64.26 | 49.48 | |
| Best-SingleCoTModel Category=Teacher Model, Protocol=SFT2026.01 | 63.2 | 50.21 | |
| ProgressiveCoTModel Category=Teacher Model, Protocol=SFT2026.01 | 62.86 | 48.45 | |
| Best-SingleCoTModel Category=Teacher Model, Protocol=few-shot2026.01 | 59.39 | 48.05 | |
| MPCoTModel Category=Teacher Model, Protocol=few-shot2026.01 | 57.51 | 45.09 |