| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| MultiArith (unseen) | UNWAVERING-FQ (+ SFT + DPO) | Average Success Count18.33 | 12 | 1mo ago | |
| StrategyQA (unseen) | Average Success Count (M.)42.65 | 12 | 1mo ago | ||
| CoinFlip (unseen) | UNWAVERING-FQ (Vicuna-7B + SFT + DPO) | Baseline Consistency Score52.2 | 3 | 1mo ago |