English Web Register Classifier
This English-specific model fine-tunes XLM-RoBERTa Large for web-register classification. It is a separate trained release from the multilingual classifier.
The trained task
The configuration contains 24 labels and omits the multilingual model's MT category. The card describes nine training epochs and an evaluated English threshold of 0.40. Broad language metadata do not change the stated English training scope.English Web Register Classifier: model card and trained artifactsEnglish Web Register Classifier: output labels and configuration
Practical assessment
Practical guidance. Compare the English and multilingual models using their own label lists and thresholds. Reusing a numeric class index from the other model can change its meaning because the output dimensions differ.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | Downloadable |
| Model identifiers | web-register-classification-en |
| Base model | XLM-RoBERTa Large.English Web Register Classifier: model card and trained artifacts |
| Output labels | 24 labels; no MT category.English Web Register Classifier: output labels and configuration |
| License | Apache 2.0.English Web Register Classifier: model card and trained artifacts |
Related models
- Multilingual Web Register Classifier
TurkuNLP ยท Web register classification
This XLM-RoBERTa Large classifier assigns hierarchical text-type labels to web documents.
Launch date not confirmedDownloadable