| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| Recall-intensive tasks suite | Xfmr++ | FDA Recall Score62.3 | 12 | 3mo ago | |
| JRT-style cloze (FDA and SWDE datasets) 1.3B / 100B checkpoints | OSDN-APF | FDA Accuracy24.1 | 2 | 20d ago | |
| Recall-intensive tasks suite (test) | - | FDA- | 0 | 3mo ago |