Reasoning on WSC Ambiguity-Augmented (200 samples)
85.2Accuracy@1DisambiguSLM
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| DisambiguSLM2026.04 | 85.2 | 88.4 | 7.8 | |
| PromptAgent2026.04 | 82.7 | 85.9 | 13.6 | |
| SPO2026.04 | 82 | 85.1 | 14.1 | |
| APE2026.04 | 80.2 | 83.1 | 15.8 | |
| OPRO2026.04 | 80.2 | 83.6 | 15.2 | |
| Rephrase2026.04 | 79.1 | 81 | 17.9 | |
| Step-back2026.04 | 78.7 | 81.3 | 17.1 | |
| TextGrad2026.04 | 78 | 80.9 | 17.2 | |
| CoT2026.04 | 77.8 | 80.4 | 18.6 | |
| PromptBreeder2026.04 | 76.7 | 79.8 | 18.4 | |
| Naïve prompting2026.04 | 72.4 | 74.1 | 21.5 |