| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| NQ (Natural Questions) | CoopRAG (GPT-4o-mini) | EM72 | 52 | 2mo ago | |
| NQ | SLEA-RL | Accuracy48.5 | 17 | 4mo ago | |
| NQ | HELP | EM43 | 13 | 5mo ago | |
| NQ (test) | Token-level F142 | 9 | 2mo ago | ||
| SimpleQuestions | GPT-4+RFKG-CoT | Accuracy0.87 | 9 | 5mo ago | |
| TriviaQA out-of-domain (test) | ECPO | Accuracy60.7 | 5 | 1mo ago | |
| NQ in-domain (test) | GiGPO | Accuracy44 | 5 | 1mo ago | |
| Assembly Knowledge Graph QA Single-hop (test) | AssemMate | Accuracy82.1 | 5 | 4mo ago |