An XML-RoBERTa based cross-lingual Sentence-BERT model distilled to cover semantic textual similarity in Finnish in addition to English. At the time of creation there were no models performing better in Finnish STS that I was aware of.
Usage instructions
This model is essentially an extended SentenceTransformer so instructions described at sbert.net apply.
The other things
The training setup, data, optimizer parameters, limitations and evaluation is described in Ch 6 here and repository.
Credit
This heavily builds on the work done by Nils Reimers et al.
Contact
Still got questions?
- Downloads last month
- 9
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.