Data Quality and Remediation Evaluation on AIC Primary Testbed
85.1F1 ScoreAIC
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| AICBackbone=gpt-4o-mini2025.12 | 85.1 | 92.7 | 78.7 | 77.8 | — | 11.7 | 0.3 | |
| AICBackbone=gpt-52025.12 | 79.7 | 100 | 66.7 | 44.4 | — | 88.9 | 6.4 | |
| AICBackbone=o4-mini2025.12 | 74.1 | 92.7 | 61.7 | 61.1 | — | 27 | 3.09 | |
| Rule-Based2025.12 | 71.4 | 92.4 | 58.2 | 55.6 | — | — | — | |
| MonolithBackbone=gpt-52025.12 | 71 | 89.5 | 58.9 | 58.3 | — | 75.9 | 4.95 |