Context-Aware Question Answering on TriState-Bench Sres
94.25EMGreedy_no_ctx
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Greedy_no_ctxModel=Llama2-13B2026.06 | 94.25 | 96.5 | |
| Greedy_no_ctxBackbone=Llama3-8B2026.06 | 91.5 | 94 | |
| Greedy_no_ctxModel=Qwen2.5-7B2026.06 | 74.25 | 81.73 | |
| Greedy_no_ctxBackbone=Mistral-7B2026.06 | 71.75 | 81.1 | |
| Simple InterpBackbone=Llama3-8B, tau=0.252026.06 | 71 | 75.83 | |
| Simple InterpModel=Llama2-13B, tau=0.252026.06 | 62.75 | 66.74 | |
| Simple InterpBackbone=Mistral-7B, tau=0.252026.06 | 45 | 60.01 | |
| Simple InterpBackbone=Llama3-8B, tau=0.52026.06 | 42.5 | 48.7 | |
| Simple InterpModel=Qwen2.5-7B, tau=0.252026.06 | 36.5 | 53.05 | |
| ARRBackbone=Llama3-8B2026.06 | 33.25 | 38.59 | |
| Simple InterpModel=Llama2-13B, tau=0.52026.06 | 29.25 | 33.25 | |
| Simple InterpBackbone=Mistral-7B, tau=0.52026.06 | 28.5 | 43.44 | |
| ARRModel=Mistral-7B-Inst2026.06 | 26.5 | 37.74 | |
| ARRModel=Llama3-8B-Inst2026.06 | 24.5 | 35.68 | |
| ARRBackbone=Mistral-7B2026.06 | 22.5 | 30.66 | |
| Simple InterpBackbone=Llama3-8B, tau=0.752026.06 | 17.75 | 24.2 | |
| ARRModel=Llama2-13B2026.06 | 17 | 23.17 | |
| ARRModel=Qwen2.5-7B2026.06 | 15.75 | 22.92 | |
| Simple InterpBackbone=Mistral-7B, tau=0.752026.06 | 15.75 | 27.14 | |
| Simple InterpModel=Qwen2.5-7B, tau=0.52026.06 | 12.5 | 29.81 | |
| Simple InterpModel=Llama2-13B, tau=0.752026.06 | 11.25 | 17.9 | |
| ARRModel=Llama2-13B-Inst2026.06 | 11 | 23.6 | |
| GreedyBackbone=Mistral-7B2026.06 | 5.25 | 12.99 | |
| GreedyBackbone=Llama3-8B2026.06 | 4.5 | 9.78 | |
| Simple InterpModel=Qwen2.5-7B, tau=0.752026.06 | 3.5 | 15.98 | |
| AdaCADBackbone=Llama3-8B2026.06 | 3.25 | 9.67 | |
| GreedyModel=Mistral-7B-Inst2026.06 | 2.5 | 13.93 | |
| AdaCADBackbone=Mistral-7B2026.06 | 2.25 | 10.06 | |
| GreedyModel=Llama2-13B2026.06 | 1.75 | 9.58 | |
| CADModel=Llama2-13B, alpha=0.252026.06 | 1.75 | 9.63 | |
| AdaCADModel=Llama2-13B2026.06 | 1.75 | 9.65 | |
| COIECDBackbone=Llama3-8B2026.06 | 1.75 | 6.29 | |
| CADBackbone=Mistral-7B, alpha=0.252026.06 | 1.75 | 9.68 | |
| CADModel=Llama2-13B, alpha=0.52026.06 | 1.5 | 9.13 | |
| CoCoAModel=Llama2-13B2026.06 | 1.5 | 9.13 | |
| AdaCADModel=Mistral-7B-Inst2026.06 | 1.5 | 12.79 | |
| CADBackbone=Llama3-8B, alpha=0.252026.06 | 1.5 | 7.68 | |
| GreedyModel=Qwen2.5-7B2026.06 | 1 | 10.04 | |
| CADModel=Llama2-13B, alpha=0.752026.06 | 1 | 8.45 | |
| COIECDModel=Llama2-13B2026.06 | 1 | 8.62 | |
| COIECDModel=Mistral-7B-Inst2026.06 | 1 | 11.83 | |
| CADBackbone=Llama3-8B, alpha=0.52026.06 | 1 | 7.67 | |
| CoCoABackbone=Llama3-8B2026.06 | 1 | 7.68 | |
| COIECDBackbone=Mistral-7B2026.06 | 1 | 8.61 | |
| AdaCADModel=Qwen2.5-7B2026.06 | 0.75 | 8.94 | |
| GreedyModel=Llama3-8B-Inst2026.06 | 0.75 | 11.1 | |
| CADBackbone=Llama3-8B, alpha=0.752026.06 | 0.75 | 7.19 | |
| CADBackbone=Llama3-8B, alpha=1.02026.06 | 0.75 | 7.11 | |
| CADBackbone=Mistral-7B, alpha=0.52026.06 | 0.75 | 8.34 | |
| CADModel=Qwen2.5-7B, alpha=0.252026.06 | 0.5 | 8.46 | |
| COIECDModel=Qwen2.5-7B2026.06 | 0.5 | 7.19 | |
| CADModel=Llama2-13B, alpha=1.02026.06 | 0.5 | 7.75 | |
| GreedyModel=Llama2-13B-Inst2026.06 | 0.5 | 8.9 | |
| CADBackbone=Mistral-7B, alpha=0.752026.06 | 0.5 | 6.75 | |
| CoCoABackbone=Mistral-7B2026.06 | 0.5 | 6.98 | |
| CADModel=Qwen2.5-7B, alpha=0.52026.06 | 0.25 | 7.24 | |
| CADModel=Qwen2.5-7B, alpha=0.752026.06 | 0.25 | 6.75 | |
| CADModel=Qwen2.5-7B, alpha=1.02026.06 | 0.25 | 6.3 | |
| CoCoAModel=Qwen2.5-7B2026.06 | 0.25 | 6.19 | |
| CADModel=Llama3-8B-Inst2026.06 | 0.25 | 9.55 | |
| COIECDModel=Llama3-8B-Inst2026.06 | 0.25 | 11.33 | |
| AdaCADModel=Llama3-8B-Inst2026.06 | 0.25 | 10.88 | |
| CoCoAModel=Llama3-8B-Inst2026.06 | 0.25 | 10.81 | |
| CADModel=Mistral-7B-Inst2026.06 | 0.25 | 10.61 | |
| ARRModel=Qwen2.5-7B-Inst2026.06 | 0.25 | 10.75 | |
| CADBackbone=Mistral-7B, alpha=1.02026.06 | 0.25 | 5.47 | |
| COIECDModel=Llama2-13B-Inst2026.06 | 0.2 | 8.7 | |
| AdaCADModel=Llama2-13B-Inst2026.06 | 0.2 | 8.4 | |
| CoCoAModel=Mistral-7B-Inst2026.06 | 0 | 5.66 | |
| GreedyModel=Qwen2.5-7B-Inst2026.06 | 0 | 7.74 | |
| CADModel=Qwen2.5-7B-Inst2026.06 | 0 | 5.32 | |
| COIECDModel=Qwen2.5-7B-Inst2026.06 | 0 | 6.99 | |
| AdaCADModel=Qwen2.5-7B-Inst2026.06 | 0 | 7.31 | |
| CoCoAModel=Qwen2.5-7B-Inst2026.06 | 0 | 6.42 | |
| CADModel=Llama2-13B-Inst2026.06 | 0 | 7.4 | |
| CoCoAModel=Llama2-13B-Inst2026.06 | 0 | 5.5 |