Want different options for adapting text embedding models to a domain?

Nils Reimers, creator of SBERT and researcher at HuggingFace, posted an interesting thread with different options regarding how to encode words you’ve never seen. Check out this thread with links to his talk and different options here.