Binary Research Idea Classification on D_point NeurIPS25 and ICLR25
75.58Acc2InnoEval
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| InnoEvalBackbone=DeepSeek-V3.22026.02 | 75.58 | 75.74 | |
| ScholarEvalBackbone=DeepSeek-V3.22026.02 | 65.44 | 65.02 | |
| InternAgentBackbone=DeepSeek-V3.22026.02 | 60.37 | 60.12 | |
| ResearchAgentBackbone=DeepSeek-V3.22026.02 | 59.48 | 56.44 | |
| GraphEvalBackbone=DeepSeek-V3.22026.02 | 54.83 | 45.71 | |
| RAGBackbone=DeepSeek-V3.22026.02 | 42.4 | 38.29 | |
| CoTBackbone=DeepSeek-V3.22026.02 | 40.55 | 35.98 |