| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| BigBench Hard Hyperbaton | PromptBreeder | Accuracy80.2 | 5 | 14d ago | |
| BigBench Hard Snarks | TextGrad | Accuracy0.554 | 5 | 14d ago | |
| BigBench Hard Disambiguation QA | ReElicit | Accuracy55.1 | 5 | 14d ago | |
| BIG-bench Hard Hyperbaton (test) | PE2 | Test Accuracy74 | 4 | 3mo ago |