| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| TyDi QA random subset of 10,000 samples | CAMAB | Log-Probability Drop0.893 | 12 | 1mo ago | |
| CNN/DM random subset of 10,000 samples | CAMAB | Log-Probability Drop1.129 | 12 | 1mo ago | |
| HotpotQA random subset of 10,000 samples | Log-Probability Drop0.024 | 12 | 1mo ago | ||
| CNN Dailymail (1000 examples) | ConCite | Log Probability Drop1.48 | 9 | 3mo ago | |
| HotpotQA distractor (val) | CAMAB | P@178 | 3 | 1mo ago |