Fetching the paper…

Abstraction via exemplars? A representational case study on lexical category inference in BERT · Around