Mathematical Word Problem Solving on GSM8K (sampled 500 test)
95AccuracySelf-consistency
Evaluation Results
| Method | Links | |
|---|---|---|
| Self-consistencyBackend Model=gpt-4o2025.10 | 95 | |
| Chain-of-Thought (CoT)Backend Model=gpt-4o2025.10 | 94 | |
| Self-askBackend Model=gpt-4o2025.10 | 94 | |
| FOR-PromptingBackend Model=gpt-4o, Host=True2025.10 | 94 | |
| FOR-PromptingBackend Model=gpt-4o, Host=False2025.10 | 94 | |
| Single-promptBackend Model=gpt-4o2025.10 | 92 |