DeBERTa
Model FamilyMicrosoftactive
Microsoft Research language-model family using disentangled attention and enhanced mask decoding.
- Jun 1, 2020
- —
- —
- Yes
- —
- 1
Models in this family
| Model | Family | Released | Modality | Open source | API | Status |
|---|---|---|---|---|---|---|
| DeBERTa V3 | DeBERTa | Nov 22, 2021 | — | Yes | — | active |
Releases
Jun 2020DeBERTa introducedResearch paper
Microsoft Research introduces DeBERTa, improving Transformer language models with disentangled attention and enhanced mask decoding.