Question Answering on NQ (EM)
72.74EMCoRect
Evaluation Results
| Method | Links | |
|---|---|---|
| CoRectModel=Qwen2.5-7b-instruct2026.02 | 72.74 | |
| CoRectModel=LLaMA3-8b-instruct2026.02 | 71.22 | |
| GreedyModel=Qwen2.5-7b-instruct2026.02 | 69.49 | |
| GreedyModel=LLaMA3-8b-instruct2026.02 | 68.72 | |
| AdaCADModel=Qwen2.5-7b-instruct2026.02 | 68.54 | |
| CADModel=LLaMA3-8b-instruct2026.02 | 68.17 | |
| AdaCADModel=LLaMA3-8b-instruct2026.02 | 67.57 | |
| CADModel=Qwen2.5-7b-instruct2026.02 | 67.28 | |
| COIECDModel=LLaMA3-8b-instruct2026.02 | 60.23 | |
| COIECDModel=Qwen2.5-7b-instruct2026.02 | 59.87 | |
| CoRectModel=Mistral-7b2026.02 | 56.31 | |
| CoRectModel=LLaMA2-13b-chat2026.02 | 49.01 | |
| CADModel=LLaMA2-13b-chat2026.02 | 46.88 | |
| GreedyModel=Mistral-7b2026.02 | 46.34 | |
| AdaCADModel=Mistral-7b2026.02 | 44.56 | |
| GreedyModel=LLaMA2-13b-chat2026.02 | 44.15 | |
| AdaCADModel=LLaMA2-13b-chat2026.02 | 43.4 | |
| COIECDModel=LLaMA2-13b-chat2026.02 | 43.11 | |
| COIECDModel=Mistral-7b2026.02 | 36.85 | |
| CADModel=Mistral-7b2026.02 | 33.5 |