Mathematical Reasoning on GSM8K (p0, p1, Delta, Sigma)
0.955p0 ScoreHuang et al. (2024)
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Huang et al. (2024)setting=intrinsic (GPT-4, round 1)2026.04 | 0.955 | 0.915 | -0.04 | -0.889 | |
| Huang et al. (2024)setting=oracle (GPT-4)2026.04 | 0.955 | 0.975 | 0.02 | 0.444 | |
| Madaan et al. (2023)setting=(Table 1, ChatGPT)2026.04 | 0.748 | 0.75 | 0.002 | 0.008 |