Fairness Evaluation on Utrecht
11.2Utrecht DPCOFT
Evaluation results
| Method | Links | ||
|---|---|---|---|
| COFTBackbone=Mistral-7B-Instruct2026.05 | 11.2 | 1 | |
| COFTBackbone=LLaMA-2-13B2026.05 | 11.8 | 1 | |
| DT-CD⋆Backbone=Mistral-7B-Instruct2026.05 | 13.6 | 2.6 | |
| DT-CD⋆Backbone=LLaMA-2-13B2026.05 | 14.1 | 2.8 | |
| DExpertsBackbone=Mistral-7B-Instruct2026.05 | 14.1 | 3.1 | |
| SDDBackbone=Mistral-7B-Instruct2026.05 | 14.6 | 3.9 | |
| DExpertsBackbone=LLaMA-2-13B2026.05 | 14.9 | 3.3 | |
| SDDBackbone=LLaMA-2-13B2026.05 | 15.3 | 4 | |
| VanillaBackbone=Mistral-7B-Instruct2026.05 | 17.3 | 5.8 | |
| VanillaBackbone=LLaMA-2-13B2026.05 | 18.4 | 5.8 |