Question Answering on TriState-Bench
64.92EMARR
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| ARRModel=Llama2-13B2026.06 | 64.92 | 69.37 | |
| ARRModel=Mistral-7B2026.06 | 64.17 | 70.28 | |
| ARRModel=Qwen2.5-7B2026.06 | 63.25 | 68.76 | |
| ARRModel=Llama3-8B2026.06 | 61.67 | 68.14 | |
| GreedyModel=Llama2-13B2026.06 | 60.67 | 65.4 | |
| AdaCADModel=Llama2-13B2026.06 | 60.08 | 64.96 | |
| GreedyModel=Mistral-7B2026.06 | 59.75 | 65.17 | |
| AdaCADModel=Mistral-7B2026.06 | 58.83 | 64.09 | |
| GreedyModel=Qwen2.5-7B2026.06 | 58.83 | 64.71 | |
| CoCoAModel=Llama2-13B2026.06 | 55.75 | 61.34 | |
| COIECDModel=Llama2-13B2026.06 | 55 | 60.59 | |
| AdaCADModel=Qwen2.5-7B2026.06 | 54.75 | 61.67 | |
| COIECDModel=Mistral-7B2026.06 | 52.92 | 57.88 | |
| CADModel=Llama2-13B2026.06 | 46.33 | 54.61 | |
| GreedyModel=Llama3-8B2026.06 | 42.92 | 51.49 | |
| COIECDModel=Qwen2.5-7B2026.06 | 41 | 51.92 | |
| AdaCADModel=Llama3-8B2026.06 | 34.42 | 44.76 | |
| CADModel=Mistral-7B2026.06 | 31.92 | 37.15 | |
| CoCoAModel=Mistral-7B2026.06 | 29.33 | 41.76 | |
| COIECDModel=Llama3-8B2026.06 | 22.33 | 34.87 | |
| CoCoAModel=Llama3-8B2026.06 | 20.75 | 33.7 | |
| CADModel=Qwen2.5-7B2026.06 | 20.75 | 37.38 | |
| CoCoAModel=Qwen2.5-7B2026.06 | 19.17 | 34.4 | |
| CADModel=Llama3-8B2026.06 | 14.58 | 28.7 |