Machine-generated text detection on MAGE squad (held-out)
97.9AUROCconfidence-weighted ensemble
Evaluation results
| Method | Links | ||
|---|---|---|---|
| confidence-weighted ensembleEnsemble Strategy=Confidence-weighted (Eq. 1)2026.07 | 97.9 | 78.8 | |
| Vanilla RoBERTa baselineBackbone=RoBERTa-large2026.07 | 97.5 | 69.2 | |
| FOMAML+LoRAK-shot=102026.07 | 96.1 | 72 |