| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Knowledge cutoff success | Factual | Unlearn Success Rate82.5 | 21 | |
| Factual Question Answering | Factual Category Average (test) | Accuracy31.38 | 18 | |
| Scene Graph Parsing | FACTUAL (test) | Completeness0.92 | 5 | |
| Factual Recall | Factual (test) | AUROC0.756 | 3 | |
| Factual Recall | factual | Tau0.734 | 3 |