Semantic Anomaly Detection on Pylkkänen and McElree Semantic
95.5AccuracyMasked language model
Evaluation results
| Method | Links | |
|---|---|---|
| Masked language modelBackbone=BERT, Anomaly Model=MLM, Layer Selection=Last layer2021.05 | 95.5 | |
| Masked language modelBackbone=RoBERTa, Anomaly Model=MLM, Layer Selection=Last layer2021.05 | 95.5 | |
| Gaussian anomaly modelBackbone=RoBERTa, Anomaly Model=GM, Layer Selection=Best layer2021.05 | 93.2 | |
| Gaussian anomaly modelBackbone=BERT, Anomaly Model=GM, Layer Selection=Best layer2021.05 | 72.7 | |
| Masked language modelBackbone=XLNet, Anomaly Model=MLM, Layer Selection=Last layer2021.05 | 72.7 | |
| Gaussian anomaly modelBackbone=XLNet, Anomaly Model=GM, Layer Selection=Best layer2021.05 | 63.6 |