Error Detection on MuSiQue
0.93F1 ScoreEnsemble-A
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Ensemble-AModel=Qwen3-8B2026.02 | 0.93 | 0.91 | 0.95 | 0.62 | |
| Ensemble-AModel=Qwen 8B2026.02 | 0.93 | — | — | 0.62 | |
| DBA-AModel=Qwen3-8B2026.02 | 0.91 | 0.92 | 0.91 | 0.65 | |
| Ensemble-IModel=Qwen3-8B2026.02 | 0.91 | 0.91 | 0.9 | 0.6 | |
| DBA-AModel=Qwen 8B2026.02 | 0.91 | — | — | 0.65 | |
| Ensemble-IModel=Qwen 8B2026.02 | 0.91 | — | — | 0.6 | |
| DBA-IModel=Qwen3-8B2026.02 | 0.89 | 0.92 | 0.87 | 0.65 | |
| DBA-IModel=Qwen 8B2026.02 | 0.89 | — | — | 0.65 | |
| Ensemble-AModel=Llama 3.3 70B2026.02 | 0.88 | 0.85 | 0.92 | 0.72 | |
| Ensemble-IModel=Llama 3.3 70B2026.02 | 0.88 | 0.85 | 0.91 | 0.7 | |
| Ensemble-AModel=Llama 3.3 70B2026.02 | 0.88 | — | — | 0.72 | |
| Ensemble-IModel=Llama 3.3 70B2026.02 | 0.88 | — | — | 0.7 | |
| DBA-AModel=Llama 3.3 70B2026.02 | 0.86 | 0.88 | 0.85 | 0.74 | |
| DBA-AModel=Llama 3.3 70B2026.02 | 0.86 | — | — | 0.74 | |
| DBA-IModel=Llama 3.3 70B2026.02 | 0.84 | 0.86 | 0.82 | 0.71 | |
| DBA-IModel=Llama 3.3 70B2026.02 | 0.84 | — | — | 0.71 | |
| AYSModel=Qwen3-8B2026.02 | 0.79 | 0.93 | 0.68 | 0.64 | |
| AYSModel=Qwen 8B2026.02 | 0.79 | — | — | 0.64 | |
| AYSModel=Llama 3.3 70B2026.02 | 0.7 | 0.9 | 0.57 | 0.69 | |
| AYSModel=Llama 3.3 70B2026.02 | 0.7 | — | — | 0.69 | |
| Ensemble-AModel=GPT 5.12026.02 | 0.69 | 0.75 | 0.65 | 0.73 | |
| Ensemble-AModel=GPT 5.12026.02 | 0.69 | — | — | 0.73 | |
| Ensemble-IModel=GPT 5.12026.02 | 0.68 | 0.69 | 0.66 | 0.71 | |
| Ensemble-IModel=GPT 5.12026.02 | 0.68 | — | — | 0.71 | |
| DBA-AModel=GPT 5.12026.02 | 0.66 | 0.75 | 0.59 | 0.72 | |
| DBA-AModel=GPT 5.12026.02 | 0.66 | — | — | 0.72 | |
| DBA-IModel=GPT 5.12026.02 | 0.64 | 0.68 | 0.6 | 0.68 | |
| DBA-IModel=GPT 5.12026.02 | 0.64 | — | — | 0.68 | |
| IC-IDKModel=Qwen3-8B2026.02 | 0.54 | 0.95 | 0.38 | 0.61 | |
| IC-IDKModel=Qwen 8B2026.02 | 0.54 | — | — | 0.62 | |
| AYSModel=GPT 5.12026.02 | 0.38 | 0.91 | 0.24 | 0.61 | |
| AYSModel=GPT 5.12026.02 | 0.38 | — | — | 0.61 | |
| IC-IDKModel=GPT 5.12026.02 | 0.27 | 0.91 | 0.16 | 0.57 | |
| IC-IDKModel=GPT 5.12026.02 | 0.27 | — | — | 0.57 | |
| IC-IDKModel=Llama 3.3 70B2026.02 | 0.22 | 0.96 | 0.12 | 0.55 | |
| IC-IDKModel=Llama 3.3 70B2026.02 | 0.22 | — | — | 0.55 |