Mathematical Reasoning on GSM8K (Source Lib., Unseen Lib., Delta)
93.21Accuracy (Source Lib.)CARD
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| CARDBase Model=GPT-4o-mini2026.06 | 93.21 | 91.44 | 1.77 | |
| SIGMABase Model=GPT-4o-mini2026.06 | 92.96 | 91.86 | 1.1 | |
| G-DesignerBase Model=GPT-4o-mini2026.06 | 89.26 | 87.28 | 1.98 | |
| GPTSwarmBase Model=GPT-4o-mini2026.06 | 87.83 | 84.47 | 3.36 | |
| LLM-DebateBase Model=GPT-4o-mini2026.06 | 87.32 | 85.29 | 2.03 |