Review
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Personalized Generation | Review (test) | Accuracy95.76 | 10 | |
| Personalized Response Generation | Review Interpolated Users | Winrate84.6 | 8 | |
| Personalized Response Generation | Review Trained Users | Winrate92.3 | 8 | |
| Abstract Screening | Review 1 821 abstracts (Final Includes) | False Positives45 | 8 | |
| Full-Text Screening | Review 1 | False Positives18 | 8 | |
| Document-Level Anomaly Detection | Review (test) | AUROC0.9594 | 7 | |
| Token-Level Anomaly Detection | Review (test) | AUROC0.8271 | 7 | |
| scoring | Review-5K | MAE1.957 | 5 | |
| Full-text inclusion screening | Review 2 (7741 abstracts) | False Positives (FP)87 | 5 | |
| Abstract Screening | Review 2 (Final Includes) | Metric- | 0 |
Showing 10 of 10 rows