Error Detection on FRAMES
95F1 ScoreEnsemble-I
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Ensemble-IModel=Qwen3-8B2026.02 | 95 | 92 | 99 | 69 | |
| Ensemble-IModel=Qwen 8B2026.02 | 95 | — | — | 69 | |
| DBA-IModel=Qwen3-8B2026.02 | 94 | 94 | 95 | 73 | |
| Ensemble-AModel=Qwen3-8B2026.02 | 94 | 92 | 96 | 67 | |
| DBA-IModel=Qwen 8B2026.02 | 94 | — | — | 73 | |
| Ensemble-AModel=Qwen 8B2026.02 | 94 | — | — | 67 | |
| DBA-AModel=Qwen3-8B2026.02 | 93 | 94 | 92 | 74 | |
| DBA-AModel=Qwen 8B2026.02 | 93 | — | — | 74 | |
| Ensemble-AModel=Llama 3.3 70B2026.02 | 90 | 85 | 95 | 75 | |
| Ensemble-IModel=Llama 3.3 70B2026.02 | 90 | 85 | 95 | 74 | |
| Ensemble-AModel=Llama 3.3 70B2026.02 | 90 | — | — | 75 | |
| Ensemble-IModel=Llama 3.3 70B2026.02 | 90 | — | — | 74 | |
| DBA-AModel=Llama 3.3 70B2026.02 | 89 | 88 | 89 | 78 | |
| DBA-IModel=Llama 3.3 70B2026.02 | 89 | 87 | 91 | 77 | |
| DBA-AModel=Llama 3.3 70B2026.02 | 89 | — | — | 78 | |
| DBA-IModel=Llama 3.3 70B2026.02 | 89 | — | — | 77 | |
| AYSModel=Qwen3-8B2026.02 | 88 | 94 | 83 | 71 | |
| AYSModel=Qwen 8B2026.02 | 88 | — | — | 71 | |
| AYSModel=Llama 3.3 70B2026.02 | 72 | 87 | 61 | 68 | |
| AYSModel=Llama 3.3 70B2026.02 | 72 | — | — | 68 | |
| DBA-IModel=GPT 5.12026.02 | 66 | 62 | 70 | 77 | |
| DBA-IModel=GPT 5.12026.02 | 66 | — | — | 77 | |
| Ensemble-IModel=GPT 5.12026.02 | 64 | 56 | 74 | 77 | |
| Ensemble-IModel=GPT 5.12026.02 | 64 | — | — | 77 | |
| Ensemble-AModel=GPT 5.12026.02 | 60 | 59 | 60 | 73 | |
| Ensemble-AModel=GPT 5.12026.02 | 60 | — | — | 73 | |
| DBA-AModel=GPT 5.12026.02 | 58 | 66 | 52 | 71 | |
| DBA-AModel=GPT 5.12026.02 | 58 | — | — | 71 | |
| AYSModel=GPT 5.12026.02 | 51 | 64 | 42 | 67 | |
| AYSModel=GPT 5.12026.02 | 51 | — | — | 67 | |
| IC-IDKModel=Qwen3-8B2026.02 | 44 | 94 | 29 | 57 | |
| IC-IDKModel=Qwen 8B2026.02 | 44 | — | — | 57 | |
| IC-IDKModel=GPT 5.12026.02 | 24 | 73 | 14 | 56 | |
| IC-IDKModel=GPT 5.12026.02 | 24 | — | — | 56 | |
| IC-IDKModel=Llama 3.3 70B2026.02 | 13 | 94 | 7 | 53 | |
| IC-IDKModel=Llama 3.3 70B2026.02 | 13 | — | — | 53 |