| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| iNatDisco-50K | DiscoPER | Recall0.6667 | 7 | 24d ago | |
| iNatDisco 800 | DiscoPER | Recall8 | 7 | 24d ago | |
| 40-task evaluation set Overall 2025 | OpenEvolve | Expert Score (ES)4.78 | 5 | 1mo ago | |
| DiscoveryBench open-ended discovery (evaluation set) | OpenEvolve | ES4.91 | 5 | 1mo ago | |
| BLADE open-ended discovery (evaluation set) | OpenEvolve | Expert Score4.8 | 5 | 1mo ago | |
| BixBench open-ended discovery (evaluation set) | OpenEvolve | ES Score4.74 | 5 | 1mo ago |