| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| Scientific Idea Generation (n=96) | EvoIdeator | Grounding0.99 | 10 | 25d ago | |
| AI Idea Bench 2025 | FlowPIE | Reward Novelty0.75 | 7 | 18d ago | |
| Scientific idea generation | EvoScientist | Novelty Win Rate96.67 | 7 | 1mo ago | |
| IdeaBench | FlowPIE | Semantic Similarity0.559 | 6 | 18d ago | |
| Scientific Ideas Human Evaluation | EvoScientist | Novelty Win Rate96.67 | 4 | 1mo ago |