Best Research Idea Selection on D_group
65.12Best ScoreInnoEval
Evaluation results
| Method | Links | |
|---|---|---|
| InnoEvalBackbone=DeepSeek-V3.22026.02 | 65.12 | |
| ScholarEvalBackbone=DeepSeek-V3.22026.02 | 49.42 | |
| InternAgentBackbone=DeepSeek-V3.22026.02 | 41.28 | |
| ResearchAgentBackbone=DeepSeek-V3.22026.02 | 40.12 | |
| CoTBackbone=DeepSeek-V3.22026.02 | 36.63 | |
| RAGBackbone=DeepSeek-V3.22026.02 | 34.3 | |
| GraphEvalBackbone=DeepSeek-V3.22026.02 | 20.93 |