Pair-wise Research Idea Comparison on D_pair Easy
80.81AcceasyInnoEval
Evaluation Results
| Method | Links | |
|---|---|---|
| InnoEvalBackbone=DeepSeek-V3.22026.02 | 80.81 | |
| ScholarEvalBackbone=DeepSeek-V3.22026.02 | 74.42 | |
| InternAgentBackbone=DeepSeek-V3.22026.02 | 62.65 | |
| ResearchAgentBackbone=DeepSeek-V3.22026.02 | 52.32 | |
| CoTBackbone=DeepSeek-V3.22026.02 | 46.51 | |
| RAGBackbone=DeepSeek-V3.22026.02 | 44.19 | |
| GraphEvalBackbone=DeepSeek-V3.22026.02 | 43.6 |