Idea Generation Assessment on AI-Idea-Bench 2025
4.44Motivation ScoreFlowPIE
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| FlowPIE2026.03 | 4.44 | — | — | — | — | — | 3.85 | |
| Initial Population2026.03 | 4.26 | — | — | — | — | — | 3.61 | |
| VirSci2026.03 | 3.95 | — | — | — | — | — | 3.84 | |
| Research Agent2026.03 | 3.78 | — | — | — | — | — | 3.45 | |
| Chain-of-Ideas2026.03 | 3.74 | — | — | — | — | — | 3.5 | |
| SCIPIP2026.03 | 3.68 | — | — | — | — | — | — | |
| FlowPIE2026.03 | 0.78 | — | — | — | — | — | 0.635 | |
| Initial Population2026.03 | 0.753 | — | — | — | — | — | 0.622 | |
| VirSci2026.03 | 0.56 | — | — | — | — | — | 0.58 | |
| Research Agent2026.03 | 0.51 | — | — | — | — | — | 0.497 | |
| Chain-of-Ideas2026.03 | 0.507 | — | — | — | — | — | 0.433 | |
| SCIPIP2026.03 | 0.464 | — | — | — | — | — | — | |
| AI-ResearcherBackbone=gemini-2.5-flash, Self-evolving memory=disabled2026.03 | — | 2.81 | 2.02 | 24.92 | 24.69 | 16.8 | — | |
| AI-ScientistBackbone=gemini-2.5-flash, Self-evolving memory=disabled2026.03 | — | 3.59 | 2.73 | 25.03 | 26.08 | 12.1 | — | |
| Chain-of-Ideas2026.03 | — | — | — | 4.63 | 3.79 | — | — | |
| FlowPIE2026.03 | — | — | — | 4.64 | 3.72 | — | — | |
| Initial Population2026.03 | — | — | — | 4.58 | 3.58 | — | — | |
| Research Agent2026.03 | — | — | — | 4.56 | 3.59 | — | — | |
| SciDERBackbone=gemini-2.5-flash, Self-evolving memory=disabled2026.03 | — | 3.78 | 3.5 | 47.06 | 44.52 | 24 | — | |
| SCIPIPBackbone=gemini-2.5-flash, Self-evolving memory=disabled2026.03 | — | 2.44 | — | 25.06 | — | — | — | |
| SCIPIP2026.03 | — | — | — | 4.18 | — | — | — | |
| VIRSCBackbone=gemini-2.5-flash, Self-evolving memory=disabled2026.03 | — | 2.94 | 2.12 | 24.87 | 24.65 | 13.3 | — | |
| VirSci2026.03 | — | — | — | 4.32 | 3.68 | — | — |