Scientific Narrative Generation on ScienceNarrative
90.25Background CreativityROUNDTABLE POLICY
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| ROUNDTABLE POLICYModel=Ours(RP)2025.09 | 90.25 | 89.95 | 91.04 | 90.39 | 91.28 | 91.18 | 90.15 | 89.86 | 90.99 | |
| GPT-o3Model=GPT-o32025.09 | 84.07 | 84.25 | 86.46 | 85.92 | 86.7 | 85.84 | 83.9 | 81.52 | 84.86 | |
| DeepSeek-chatModel=DeepSeek-chat2025.09 | 84.05 | 83.74 | 88.28 | 84.21 | 84.58 | 86.08 | 83.68 | 81.56 | 86.5 | |
| GPT-o3-mModel=GPT-o3-m2025.09 | 82.65 | 84.78 | 87.26 | 83.15 | 86.86 | 86.44 | 81.5 | 81.82 | 85.15 | |
| Claude-3.7SModel=Claude-3.7S2025.09 | 82.47 | 84.73 | 88.17 | 85.52 | 86.65 | 86.89 | 84.53 | 82.13 | 86.41 | |
| Grok-3Model=Grok-32025.09 | 78.5 | 79.1 | 85.5 | 80.96 | 81.26 | 83.58 | 78.81 | 76.34 | 83.04 | |
| GPT-4.1nModel=GPT-4.1n2025.09 | 77.39 | 80.94 | 85.59 | 80.48 | 82.3 | 83.97 | 77.97 | 77.22 | 82.94 | |
| GPT-4.1Model=GPT-4.12025.09 | 76.96 | 81.37 | 86.19 | 81.65 | 83.19 | 84.75 | 78.81 | 77.91 | 83.9 | |
| GPT-4o-mModel=GPT-4o-m2025.09 | 72.78 | 74.64 | 82.19 | 75.44 | 76.18 | 79.81 | 72.78 | 70.69 | 78.82 | |
| GPT-4oModel=GPT-4o2025.09 | 70.99 | 74.73 | 82.75 | 75.49 | 74.81 | 79.63 | 72.15 | 70.53 | 79.64 |