| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| TOFU author-profile questions (held-out) | FINCH | Task Accuracy84.3 | 22 | 14d ago | |
| Knowledge Acquisition | Task Accuracy83.5 | 10 | 14d ago | ||
| LawQA | ROUGE-144.71 | 7 | 1mo ago | ||
| VideoMMMU | SiLVR | Delta Knowledge17.2 | 5 | 1mo ago | |
| Wikipedia Knowledge Acquisition Out-of-distribution (OOD) | Accuracy100 | 5 | 3mo ago | ||
| Wikipedia Knowledge Acquisition In-distribution (test) | Accuracy (strict)91 | 5 | 3mo ago | ||
| Newly learned facts | SAT | Base Accuracy93 | 4 | 3mo ago | |
| BioQA | GTKA | Accuracy4.65 | 2 | 1mo ago |