Research Idea Generation on Expert Evaluation Education Domain (test)
80Novelty Winrate vs QwenDeepInnovator
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| DeepInnovatorevaluation_mode=Human Expert Annotation2026.02 | 80 | 60 | 57.1 | 0 | 53.3 | 0 | 75 | 36.4 |
Loading the SOTA2 catalog…