Algebraic Word Problems on AQuA (Source/Unseen Generalization)
83.67Source Library AccuracySIGMA
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| SIGMABase Model=GPT-4o-mini2026.06 | 83.67 | 82.5 | 1.17 | |
| CARDBase Model=GPT-4o-mini2026.06 | 80.41 | 76.85 | 3.56 | |
| G-DesignerBase Model=GPT-4o-mini2026.06 | 76.6 | 70.3 | 6.3 | |
| GPTSwarmBase Model=GPT-4o-mini2026.06 | 74.65 | 69.14 | 5.51 | |
| LLM-DebateBase Model=GPT-4o-mini2026.06 | 72.86 | 66.96 | 5.9 |