Question Answering on TabMWP
70.6EMCoRect
Evaluation Results
| Method | Links | |
|---|---|---|
| CoRectModel=Qwen2.5-7b-instruct2026.02 | 70.6 | |
| COIECDModel=Qwen2.5-7b-instruct2026.02 | 64.35 | |
| AdaCADModel=Qwen2.5-7b-instruct2026.02 | 60 | |
| GreedyModel=Qwen2.5-7b-instruct2026.02 | 59.9 | |
| CADModel=Qwen2.5-7b-instruct2026.02 | 58.9 | |
| CoRectModel=LLaMA3-8b-instruct2026.02 | 52.6 | |
| COIECDModel=LLaMA3-8b-instruct2026.02 | 51.95 | |
| CADModel=LLaMA3-8b-instruct2026.02 | 48.9 | |
| GreedyModel=LLaMA3-8b-instruct2026.02 | 47.4 | |
| AdaCADModel=LLaMA3-8b-instruct2026.02 | 47.3 | |
| CoRectModel=Mistral-7b2026.02 | 38.5 | |
| COIECDModel=Mistral-7b2026.02 | 36.5 | |
| CADModel=Mistral-7b2026.02 | 35.4 | |
| CADModel=LLaMA2-13b-chat2026.02 | 27.3 | |
| CoRectModel=LLaMA2-13b-chat2026.02 | 27.2 | |
| COIECDModel=LLaMA2-13b-chat2026.02 | 26.7 | |
| AdaCADModel=Mistral-7b2026.02 | 25.4 | |
| GreedyModel=Mistral-7b2026.02 | 25.2 | |
| GreedyModel=LLaMA2-13b-chat2026.02 | 17.9 | |
| AdaCADModel=LLaMA2-13b-chat2026.02 | 17.7 |