| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| TriviaQA | LLM-PeerReview-W | Accuracy77 | 16 | 4mo ago | |
| MMLU | OLMo 3 | MMLU Factual Recall Accuracy58 | 14 | 4mo ago | |
| Factual recall GPT-2 small (n=500) | IGSD | Top-1 Accuracy81.6 | 12 | 1mo ago | |
| Tri-REx (test) | Hit@133.9 | 6 | 2mo ago | ||
| Factual (test) | Post-solution probe | AUROC0.756 | 3 | 16d ago | |
| factual | HVP | Tau0.734 | 3 | 1mo ago |