Scientific Research Paper Generation Evaluation on ML Topics (ML01, ML02, ML12, ML16, ML18, ML20)
5.67Overall ScoreSAGE
Evaluation results
| Method | Links | |
|---|---|---|
| SAGEevaluation_type=Blind human evaluation by ML-PhD reviewers, averaging=Mean over topics and reviewers2026.06 | 5.67 | |
| AI-Scientist-v2evaluation_type=Blind human evaluation by ML-PhD reviewers, averaging=Mean over topics and reviewers2026.06 | 4.72 | |
| AutoResearchClawevaluation_type=Blind human evaluation by ML-PhD reviewers, averaging=Mean over topics and reviewers2026.06 | 3 |