'Sorry, I didn't understand that'

This paper proposes WFST-BERT model which augments the fine-tuning of BERT-like architecture with weighted finite-state transducer to

Discover how BERT revolutionizes Conversational AI slot filling, enhancing user intent recognition and response accuracy

bert slot Thread by @ogrisel: I prepared a new notebook for my Deep Learning class: Joint Intent Classification and Slot Filling with BERT: orggithub

sugar supreme powernudge slot filling tasks with the separate BERT embedding The model joint intent classification and slot filling arXiv preprint arXiv