Arithmetic Reasoning on GSM Reversed
90.3AccuracyGPT-4o
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4oPrompting=CoT2025.12 | 90.3 | |
| GPT-4oPrompting=decl. PL-1S2025.12 | 87.8 | |
| GPT-4oPrompting=PY-ZS2025.12 | 87.6 | |
| GPT-4oPrompting=decl. PY-1S2025.12 | 87.5 | |
| GPT-4oPrompting=PL-ZS2025.12 | 76.2 | |
| CodeLlama13BPrompting=decl. PL-1S2025.12 | 34.1 | |
| CodeLlama13BPrompting=decl. PY-1S2025.12 | 29.1 |