State Value Generation with Prompt Learning and Self-Training for Low-Resource Dialogue State Tracking

Ming Gu, Yan Yang, Chengcai Chen, Zhou Yu
Proceedings of the 15th Asian Conference on Machine Learning, PMLR 222:390-405, 2024.

Abstract

Recently, low-resource dialogue state tracking (DST) has received increasing attention. First obtaining state values then based on values to generate slot types has made great progress in this task. However, obtaining state values is still an under-studied problem. Existing extraction-based approaches cannot capture values that require the understanding of context and are not generalizable either. To address these issues, we propose a novel State VAlue Generation based framework (SVAG), decomposing DST into state value generation and domain slot generation. Specifically, we propose to generate state values and use self-training to further improve state value generation. Moreover, we design an estimator aiming at detecting incomplete generation and incorrect generation for pseudo-labeled data selection during self-training. Experimental results on the MultiWOZ 2.1 dataset show that our method which has only less than 1 billion parameters achieves state-of-the-art performance under the data ratio settings of 5%, 10%, and 25% when limited to models under 100 billion parameters. Compared to models with more than 100 billion parameters, SVAG still reaches competitive results.

Cite this Paper

BibTeX
@InProceedings{pmlr-v222-gu24a, title = {State Value Generation with Prompt Learning and Self-Training for Low-Resource Dialogue State Tracking}, author = {Gu, Ming and Yang, Yan and Chen, Chengcai and Yu, Zhou}, booktitle = {Proceedings of the 15th Asian Conference on Machine Learning}, pages = {390--405}, year = {2024}, editor = {Yanıkoğlu, Berrin and Buntine, Wray}, volume = {222}, series = {Proceedings of Machine Learning Research}, month = {11--14 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v222/gu24a/gu24a.pdf}, url = {https://proceedings.mlr.press/v222/gu24a.html}, abstract = {Recently, low-resource dialogue state tracking (DST) has received increasing attention. First obtaining state values then based on values to generate slot types has made great progress in this task. However, obtaining state values is still an under-studied problem. Existing extraction-based approaches cannot capture values that require the understanding of context and are not generalizable either. To address these issues, we propose a novel State VAlue Generation based framework (SVAG), decomposing DST into state value generation and domain slot generation. Specifically, we propose to generate state values and use self-training to further improve state value generation. Moreover, we design an estimator aiming at detecting incomplete generation and incorrect generation for pseudo-labeled data selection during self-training. Experimental results on the MultiWOZ 2.1 dataset show that our method which has only less than 1 billion parameters achieves state-of-the-art performance under the data ratio settings of 5%, 10%, and 25% when limited to models under 100 billion parameters. Compared to models with more than 100 billion parameters, SVAG still reaches competitive results.} }
Endnote
%0 Conference Paper %T State Value Generation with Prompt Learning and Self-Training for Low-Resource Dialogue State Tracking %A Ming Gu %A Yan Yang %A Chengcai Chen %A Zhou Yu %B Proceedings of the 15th Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Berrin Yanıkoğlu %E Wray Buntine %F pmlr-v222-gu24a %I PMLR %P 390--405 %U https://proceedings.mlr.press/v222/gu24a.html %V 222 %X Recently, low-resource dialogue state tracking (DST) has received increasing attention. First obtaining state values then based on values to generate slot types has made great progress in this task. However, obtaining state values is still an under-studied problem. Existing extraction-based approaches cannot capture values that require the understanding of context and are not generalizable either. To address these issues, we propose a novel State VAlue Generation based framework (SVAG), decomposing DST into state value generation and domain slot generation. Specifically, we propose to generate state values and use self-training to further improve state value generation. Moreover, we design an estimator aiming at detecting incomplete generation and incorrect generation for pseudo-labeled data selection during self-training. Experimental results on the MultiWOZ 2.1 dataset show that our method which has only less than 1 billion parameters achieves state-of-the-art performance under the data ratio settings of 5%, 10%, and 25% when limited to models under 100 billion parameters. Compared to models with more than 100 billion parameters, SVAG still reaches competitive results.
APA
Gu, M., Yang, Y., Chen, C. & Yu, Z.. (2024). State Value Generation with Prompt Learning and Self-Training for Low-Resource Dialogue State Tracking. Proceedings of the 15th Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 222:390-405 Available from https://proceedings.mlr.press/v222/gu24a.html.

Related Material