Natural language understanding typically maps single utterances to a dual level semantic frame, sentence level intent and slot labels at the word level. The best performing models force explicit interaction between intent detection and slot filling. We present a novel tri-level joint natural language understanding approach, adding domain, and explicitly exchange semantic information between all levels. This approach enables the use of multi-turn datasets which are a more natural conversational environment than single utterance. We evaluate our model on two multi-turn datasets for which we are the first to conduct joint slot-filling and intent detection. Our model outperforms state-of-the-art joint models in slot filling and intent detection on multi-turn data sets. We provide an analysis of explicit interaction locations between the layers. We conclude that including domain information improves model performance.
@article{arxiv.2305.17729,
title = {Tri-level Joint Natural Language Understanding for Multi-turn Conversational Datasets},
author = {Henry Weld and Sijia Hu and Siqu Long and Josiah Poon and Soyeon Caren Han},
journal= {arXiv preprint arXiv:2305.17729},
year = {2023}
}