ONENET: Joint domain, intent, slot prediction for spoken language understanding

By Young-Bum Kim, Sungjin Lee, Karl Stratos. 2017

75 citations recorded by the source corpus.

In practice, most spoken language understanding systems process user input in a pipelined manner; first domain is predicted, then intent and semantic slots are inferred according to the semantic frames of the predicted domain. The pipeline approach, however, has some disadvantages: error propagation and lack of information sharing. To address these issues, we present a unified neural network that jointly performs domain, intent, and slot predictions. Our approach adopts a principled architecture for multitask learning to fold in the state-of-the-art models for each task. With a few more ingredients, e.g. orthography-sensitive input encoding and curriculum training, our model delivered significant improvements in all three tasks across all domains over strong baselines, including one using oracle prediction for domain detection, on real user data of a commercial personal assistant.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Towards Open Intent Discovery for Conversational Text (2019, methodology): “Recent approaches have drawn upon progress in CNN, RNN, and CRF based language models to improve intent detection [11, 22, 24, 31, 36, 49, 52].”
Efficient Large-Scale Domain Classification with Personalized Attention (2018, background): “…with pre-training (Kim et al., 2015b; Celiky-ilmaz et al., 2016; Kim et al., 2017e), multi-task learning (Zhang and Wang, 2016; Liu and Lane, 2016; Kim et al., 2017b), transfer learning (El-Kahky et al., 2014; Kim et al., 2015a,c; Chen et al., 2016a; Yang et al., 2017), domain adaptation (Kim et…”
Iterative Delexicalization for Improved Spoken Language Understanding (2019, methodology): “Many slot values (e.g. artist, song titles) also experience a distribution shift over time, and may require sophisticated models and retraining to adapt to such changes [Kim et al., 2017b].”
End-to-End Neural Pipeline for Goal-Oriented Dialogue Systems using GPT-2 (2020, methodology): “∗ : Equal contribution The traditional approach to building a goaloriented dialogue system mostly adopts a pipelined modular architecture, with the natural language understanding (NLU) module (Kim et al., 2017; Lee et al., 2019b) that first recognizes and comprehends user’s intent and extracts value”
Joint Learning of Domain Classification and Out-of-Domain Detection with Dynamic Class Weighting for Satisficing False Acceptance Rates (2018, background): “For example, joint training of intent detection and slot-filling [8], joint training of all the three SLU components [9, 10], and joint training of multiple domains [11, 12, 13] have been shown synergistic since the jointly trained components are highly related to each other.”