Error Detection on MuSiQue (val)
1PrecisionIC-IDK
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| IC-IDKBase Model=Llama 3.1 8B2026.02 | 1 | 0.1 | 0.19 | 0.55 | |
| DBA-ABase Model=Llama 3.1 8B2026.02 | 0.96 | 0.93 | 0.95 | 0.77 | |
| AYSBase Model=Llama 3.1 8B2026.02 | 0.95 | 0.45 | 0.61 | 0.61 | |
| Ensemble-ABase Model=Llama 3.1 8B2026.02 | 0.95 | 0.96 | 0.95 | 0.67 | |
| DBA-IBase Model=Llama 3.1 8B2026.02 | 0.94 | 0.91 | 0.92 | 0.67 | |
| AYSBase Model=Qwen3-32B2026.02 | 0.94 | 0.64 | 0.76 | 0.71 | |
| IC-IDKBase Model=Qwen3-32B2026.02 | 0.94 | 0.2 | 0.33 | 0.57 | |
| DBA-ABase Model=Mistral 7B2026.02 | 0.93 | 0.92 | 0.93 | 0.72 | |
| Ensemble-IBase Model=Llama 3.1 8B2026.02 | 0.93 | 0.93 | 0.93 | 0.64 | |
| DBA-ABase Model=Qwen3-32B2026.02 | 0.93 | 0.9 | 0.92 | 0.78 | |
| IC-IDKBase Model=Gemini Pro2026.02 | 0.93 | 0.18 | 0.29 | 0.58 | |
| AYSBase Model=Mistral 7B2026.02 | 0.92 | 0.42 | 0.57 | 0.58 | |
| IC-IDKBase Model=Mistral 7B2026.02 | 0.92 | 0.58 | 0.71 | 0.61 | |
| DBA-IBase Model=Mistral 7B2026.02 | 0.92 | 0.92 | 0.92 | 0.68 | |
| Ensemble-ABase Model=Mistral 7B2026.02 | 0.92 | 0.94 | 0.93 | 0.67 | |
| Ensemble-ABase Model=Qwen3-32B2026.02 | 0.92 | 0.93 | 0.92 | 0.74 | |
| IC-IDKBase Model=Qwen 2.5 72B2026.02 | 0.92 | 0.42 | 0.58 | 0.65 | |
| DBA-IBase Model=Qwen3-32B2026.02 | 0.91 | 0.86 | 0.88 | 0.72 | |
| Ensemble-IBase Model=Qwen3-32B2026.02 | 0.91 | 0.9 | 0.9 | 0.7 | |
| Ensemble-IBase Model=Mistral 7B2026.02 | 0.9 | 0.94 | 0.92 | 0.61 | |
| DBA-ABase Model=Qwen 2.5 72B2026.02 | 0.9 | 0.83 | 0.86 | 0.76 | |
| IC-IDKBase Model=Gemini Flash2026.02 | 0.87 | 0.27 | 0.41 | 0.6 | |
| DBA-IBase Model=Qwen 2.5 72B2026.02 | 0.86 | 0.86 | 0.86 | 0.71 | |
| AYSBase Model=Gemini Flash2026.02 | 0.86 | 0.27 | 0.41 | 0.6 | |
| DBA-ABase Model=Gemini Flash2026.02 | 0.84 | 0.71 | 0.77 | 0.74 | |
| Ensemble-ABase Model=Qwen 2.5 72B2026.02 | 0.83 | 0.93 | 0.88 | 0.66 | |
| Ensemble-ABase Model=Gemini Flash2026.02 | 0.83 | 0.75 | 0.79 | 0.74 | |
| DBA-ABase Model=Gemini Pro2026.02 | 0.83 | 0.66 | 0.73 | 0.75 | |
| AYSBase Model=Qwen 2.5 72B2026.02 | 0.82 | 0.74 | 0.78 | 0.6 | |
| DBA-IBase Model=Gemini Flash2026.02 | 0.82 | 0.74 | 0.78 | 0.73 | |
| Ensemble-IBase Model=Qwen 2.5 72B2026.02 | 0.81 | 0.95 | 0.88 | 0.61 | |
| Ensemble-IBase Model=Gemini Flash2026.02 | 0.81 | 0.78 | 0.79 | 0.73 | |
| Ensemble-ABase Model=Gemini Pro2026.02 | 0.79 | 0.7 | 0.74 | 0.75 | |
| AYSBase Model=Gemini Pro2026.02 | 0.74 | 0.21 | 0.33 | 0.56 | |
| DBA-IBase Model=Gemini Pro2026.02 | 0.72 | 0.61 | 0.66 | 0.68 | |
| Ensemble-IBase Model=Gemini Pro2026.02 | 0.72 | 0.67 | 0.69 | 0.69 |