Error Detection on CRAG
91F1 ScoreEnsemble-A
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Ensemble-AModel=Qwen3-8B2026.02 | 91 | 87 | 95 | 67 | |
| Ensemble-IModel=Qwen3-8B2026.02 | 91 | 88 | 95 | 70 | |
| Ensemble-AModel=Qwen 8B2026.02 | 91 | — | — | 67 | |
| Ensemble-IModel=Qwen 8B2026.02 | 91 | — | — | 70 | |
| DBA-AModel=Qwen3-8B2026.02 | 89 | 89 | 89 | 73 | |
| DBA-IModel=Qwen3-8B2026.02 | 89 | 89 | 89 | 73 | |
| DBA-AModel=Qwen 8B2026.02 | 89 | — | — | 73 | |
| DBA-IModel=Qwen 8B2026.02 | 89 | — | — | 73 | |
| Ensemble-IModel=Llama 3.3 70B2026.02 | 77 | 80 | 74 | 74 | |
| Ensemble-IModel=Llama 3.3 70B2026.02 | 77 | — | — | 74 | |
| IC-IDKModel=Qwen3-8B2026.02 | 76 | 86 | 68 | 61 | |
| IC-IDKModel=Qwen 8B2026.02 | 76 | — | — | 61 | |
| Ensemble-AModel=GPT 5.12026.02 | 75 | 73 | 77 | 81 | |
| IC-IDKModel=Llama 3.3 70B2026.02 | 75 | 70 | 81 | 67 | |
| AYSModel=Qwen3-8B2026.02 | 75 | 90 | 64 | 68 | |
| Ensemble-AModel=GPT 5.12026.02 | 75 | — | — | 81 | |
| IC-IDKModel=Llama 3.3 70B2026.02 | 75 | — | — | 67 | |
| AYSModel=Qwen 8B2026.02 | 75 | — | — | 68 | |
| Ensemble-IModel=GPT 5.12026.02 | 74 | 73 | 75 | 80 | |
| DBA-IModel=Llama 3.3 70B2026.02 | 74 | 87 | 65 | 76 | |
| Ensemble-AModel=Llama 3.3 70B2026.02 | 74 | 76 | 72 | 70 | |
| Ensemble-IModel=GPT 5.12026.02 | 74 | — | — | 80 | |
| DBA-IModel=Llama 3.3 70B2026.02 | 74 | — | — | 76 | |
| Ensemble-AModel=Llama 3.3 70B2026.02 | 74 | — | — | 70 | |
| DBA-AModel=Llama 3.3 70B2026.02 | 71 | 79 | 65 | 71 | |
| DBA-AModel=Llama 3.3 70B2026.02 | 71 | — | — | 71 | |
| AYSModel=GPT 5.12026.02 | 66 | 75 | 59 | 74 | |
| AYSModel=GPT 5.12026.02 | 66 | — | — | 74 | |
| IC-IDKModel=GPT 5.12026.02 | 60 | 71 | 52 | 70 | |
| IC-IDKModel=GPT 5.12026.02 | 60 | — | — | 70 | |
| DBA-AModel=GPT 5.12026.02 | 53 | 77 | 40 | 67 | |
| AYSModel=Llama 3.3 70B2026.02 | 53 | 82 | 39 | 64 | |
| DBA-AModel=GPT 5.12026.02 | 53 | — | — | 67 | |
| AYSModel=Llama 3.3 70B2026.02 | 53 | — | — | 64 | |
| DBA-IModel=GPT 5.12026.02 | 49 | 75 | 37 | 65 | |
| DBA-IModel=GPT 5.12026.02 | 49 | — | — | 65 |