Abstract
While dialogue state tracking by generation-based approaches allows for better scalability and generalization, they suffer from two major limitations. First, most generation-based models adopt a multi-task learning framework that may cause gradient conflicts and low training efficiency. Second, since the dialogue state of the previous turn is usually taken as an input for the current turn, there exists inconsistency between training and inference, which is identified as turn-level exposure bias. To address the first limitation, we propose the idea of state-transition sequence and transform multi-task learning into a single generation task. To alleviate turn-level exposure bias, we propose a slot-perturb strategy to reduce the over-reliance on the previous dialogue state. Experimental results show that our method achieves a new state of the art on the MultiWOZ 2.4 dataset and performs competitively on MultiWOZ 2.1. Besides, we demonstrate that the unified generation framework with slot-perturb improves the convergence speed and relieves error accumulation.
Z. Lin and B. Guo are co-first authors and contribute equally to this work.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Budzianowski, P., et al.: MultiWOZ - a large-scale multi-domain Wizard-of-Oz dataset for task-oriented dialogue modelling. In: Proceedings of EMNLP (2018)
Chen, H., Liu, X., Yin, D., Tang, J.: A survey on dialogue systems: recent advances and new frontiers. SIGKDD Explor. Newsl. 19(2), 25–35 (2017)
Chen, J., Zhang, R., Mao, Y., Xu, J.: Parallel interactive networks for multi-domain dialogue state generation. In: Proceedings of EMNLP (2020)
Devlin, J., Chang, M.W., Lee, K., Toutanova, K.: BERT: pre-training of deep bidirectional transformers for language understanding. In: Proceedings of NAACL-HLT, Volume 1 (Long and Short Papers) (2019)
Eric, M., et al.: MultiWOZ 2.1: a consolidated multi-domain dialogue dataset with state corrections and state tracking baselines. In: Proceedings of LREC, pp. 422–428 (2020)
Feng, Y., Wang, Y., Li, H.: A sequence-to-sequence approach to dialogue state tracking. In: Proceedings of ACL/IJCNLP (2021)
Heck, M., et al.: TripPy: a triple copy strategy for value independent neural dialog state tracking. In: Proceedings of SIGDIAL (2020)
Kim, S., Yang, S., Kim, G., Lee, S.W.: Efficient dialogue state tracking by selectively overwriting memory. In: Proceedings of ACL (2020)
Lee, C.H., Cheng, H., Ostendorf, M.: Dialogue state tracking with a language model using schema-driven prompting. In: Proceedings of EMNLP, pp. 4937–4949 (2021)
Lee, H., Lee, J., Kim, T.Y.: SUMBT: slot-utterance matching for universal and scalable belief tracking. In: Proceedings of ACL, pp. 5478–5483 (2019)
Mehri, S., Eric, M., Hakkani-Tur, D.: DialoGLUE: a natural language understanding benchmark for task-oriented dialogue. arXiv e-prints arXiv:2009.13570 (2020)
Peng, B., Li, C., Li, J., Shayandeh, S., Liden, L., Gao, J.: SOLOIST: building task bots at scale with transfer learning and machine teaching. Trans. Assoc. Comput. Linguis. 9, 807–824 (2021)
Ranzato, M., Chopra, S., Auli, M., Zaremba, W.: Sequence level training with recurrent neural networks. arXiv e-prints arXiv:1511.06732 (2015)
Schmidt, F.: Generalization in generation: a closer look at exposure bias. In: Proceedings of the 3rd Workshop on Neural Generation and Translation, pp. 157–167 (2019)
Su, Y., et al.: Multi-task pre-training for plug-and-play task-oriented dialogue system. In: Proceedings of ACL, pp. 4661–4676 (2022)
Vandenhende, S., Georgoulis, S., Van Gansbeke, W., Proesmans, M., Dai, D., Van Gool, L.: Multi-task learning for dense prediction tasks: a survey. IEEE Trans. Pattern Anal. Mach. Intell. 44(7), 3614–3633 (2022)
Wang, Y., Guo, Y., Zhu, S.: Slot attention with value normalization for multi-domain dialogue state tracking. In: Proceedings of EMNLP (2020)
Wu, C.S., Madotto, A., Hosseini-Asl, E., Xiong, C., Socher, R., Fung, P.: Transferable multi-domain state generator for task-oriented dialogue systems. In: Proceedings of ACL, pp. 808–819 (2019)
Ye, F., Manotumruksa, J., Yilmaz, E.: MultiWOZ 2.4: a multi-domain task-oriented dialogue dataset with essential annotation corrections to improve state tracking evaluation. CoRR abs/2104.00773 (2021)
Ye, F., Manotumruksa, J., Zhang, Q., Li, S., Yilmaz, E.: Slot self-attentive dialogue state tracking. In: Proceedings of WWW, pp. 1598–1608 (2021)
Zeng, Y., Nie, J.Y.: Multi-domain dialogue state tracking-a purely transformer-based generative approach. CoRR abs/2010.14061 (2020)
Zhang, J., et al.: Find or classify? Dual strategy for slot-value predictions on multi-domain dialog state tracking. In: Proceedings of COLING, pp. 154–167 (2020)
Zhao, J., Mahdieh, M., Zhang, Y., Cao, Y., Wu, Y.: Effective sequence-to-sequence dialogue state tracking. In: Proceedings of EMNLP, pp. 7486–7493 (2021)
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Lin, Z., Guo, B., Shi, T., Li, Y., Quan, X., Li, L. (2023). A Unified Generation Approach for Robust Dialogue State Tracking. In: Liu, F., Duan, N., Xu, Q., Hong, Y. (eds) Natural Language Processing and Chinese Computing. NLPCC 2023. Lecture Notes in Computer Science(), vol 14302. Springer, Cham. https://doi.org/10.1007/978-3-031-44693-1_61
Download citation
DOI: https://doi.org/10.1007/978-3-031-44693-1_61
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-44692-4
Online ISBN: 978-3-031-44693-1
eBook Packages: Computer ScienceComputer Science (R0)