References & Citations
Computer Science > Computation and Language
Title: Multi-Domain Dialogue State Tracking -- A Purely Transformer-Based Generative Approach
(Submitted on 24 Oct 2020 (this version), latest version 8 Apr 2021 (v2))
Abstract: We investigate the problem of multi-domain Dialogue State Tracking (DST) with open vocabulary. Existing approaches exploit BERT encoder and copy-based RNN decoder, where the encoder first predicts the state operation, and then the decoder generates new slot values. However, in this stacked encoder-decoder structure, the operation prediction objective only affects the BERT encoder and the value generation objective mainly affects the RNN decoder. In this paper, we propose a purely Transformer-based framework that uses BERT as both encoder and decoder. In so doing, the operation prediction objective and the value generation objective can jointly optimize our model for DST. At the decoding step, we re-use the hidden states of the encoder in the self-attention mechanism of the corresponding decoder layer to construct a flat model structure for effective parameter updating. Experimental results show that our approach substantially outperforms the existing state-of-the-art framework, and it also achieves very competitive performance to the best ontology-based approaches.
Submission history
From: Yan Zeng [view email][v1] Sat, 24 Oct 2020 04:54:52 GMT (8551kb,D)
[v2] Thu, 8 Apr 2021 02:04:05 GMT (1568kb,D)
Link back to: arXiv, form interface, contact.