Research Automation on Three real research tasks Human researcher evaluation
9.333AlignmentNanoResearch
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| NanoResearchRound=12026.05 | 9.333 | 6 | 1 | 64.66 | 7 | |
| NanoResearchRound=22026.05 | 9.333 | 7 | 1 | 85.02 | 8 | |
| NanoResearchRound=32026.05 | 9.333 | 6.667 | 1 | 86.03 | 7.667 | |
| DeepScientist2026.05 | 6.333 | 5 | 1 | 60.94 | 5.333 | |
| EvoScientist2026.05 | 6 | 4.667 | 1 | 65.37 | 4 | |
| AI Scientist-v22026.05 | 5.333 | 4 | 1 | 49.65 | 4.333 | |
| AI-Researcher2026.05 | 4.333 | 3.333 | 1 | 54.95 | 4.667 |