| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| FeedSum | Gemini | Consistency Score96 | 11 | 3mo ago | |
| 100 articles (Protocol 1) | Llama-70B Faithfulness (%)19 | 6 | 1mo ago | ||
| CNN/DM Within AF (test) | Adversarial | Pairwise Ranking Accuracy59.2 | 4 | 4mo ago | |
| CNN/DM All Pairs (test) | Adversarial | Pairwise Ranking Accuracy85.27 | 4 | 4mo ago |