Knowledge-Grounded Dialogue Generation with Contrastive Knowledge Selection
Authors: 
Bin Wang, Fuyong Xu, Zhenfang Zhu, Peiyu LiuAuthors Info & Claims
Bin Wang, Fuyong Xu, Zhenfang Zhu, Peiyu LiuAuthors Info & ClaimsPages 710 - 719
Abstract
Knowledge selection is the key component in knowledge-ground dialogues, which aims to choice correct knowledge based on external knowledge for dialogue generation. The quality of knowledge selection depend on knowledge representation methods. However, the knowledge representation exploration is still challenging. We propose a knowledge-grounded dialogue model, which incorporates a knowledge-grounded module and a dialogue generation module, aiming to choose the most appropriate knowledge and fuse it into response generation. In addition, supervised contrastive knowledge representation signal is designed to obtain knowledge representation dynamically. Experiments on FoCus dataset show that our model outperforms the baseline models. The ablation study further demonstrates the effectiveness of each sub-modules.
References
[1]
Zhang, S., et al.: Personalizing dialogue agents: i have a dog, do you have pets too?. In: ACL (Volume 1: Long Papers) (2018)
[2]
Kim, M., et al.: Dual task framework for improving persona-grounded dialogue dataset. In: AAAI, vol. 36, no. 10 (2022)
[3]
Brown, T., et al.: Language models are few-shot learners. In: NeurIPS, vol. 33, pp. 1877–1901 (2020)
[4]
Raffel C et al. Exploring the limits of transfer learning with a unified text-to-text transformer J. Mach. Learn. Res. 2020 21 1 5485-5551
[5]
Wang, P., et al.: OFA: unifying architectures, tasks, and modalities through a simple sequence-to-sequence learning framework. In: International Conference on Machine Learning. PMLR (2022)
[6]
Dinan, E., et al.: Wizard of Wikipedia: knowledge-powered conversational agents. In: ICLR (2018)
[7]
Smith, E.M., et al.: Can you put it all together: evaluating conversational agents’ ability to blend skills. In: ACL (2020)
[8]
Jang, Y., et al.: Call for customized conversation: customized conversation grounding persona and knowledge. In: AAAI, vol. 36, no. 10 (2022)
[9]
Saha, S., Das, S., Srihari, R.K.: Proto-gen: an end-to-end neural generator for persona and knowledge grounded response generation. In: Proceedings of the 1st Workshop on Customized Chat Grounding Persona and Knowledge (2022)
[10]
Lee, Y.-J., et al.: PERSONACHATGEN: generating Personalized Dialogues using GPT-3. In: Proceedings of the 1st Workshop on Customized Chat Grounding Persona and Knowledge (2022)
[11]
Bak, J., Oh, A.: Variational hierarchical user-based conversation model. In: EMNLP-IJCNLP (2019)
[12]
Nakamura, K., et al.: HybriDialogue: an information-seeking dialogue dataset grounded on tabular and textual data. In: Findings of the Association for Computational Linguistics: ACL (2022)
[13]
Xu, F., et al.: Diverse dialogue generation by fusing mutual persona-aware and self-transferrer. In: APIN, pp. 1–14 (2022)
[14]
Wang, Y., et al.: Improving persona understanding for persona-based dialogue generation with diverse knowledge selection. In: ICPR. IEEE Computer Society (2022)
[15]
Smith, E.M., et al.: Can you put it all together: evaluating conversational agents’ ability to blend skills. In: ACL (2020)
[16]
Chen, T., et al.: A simple framework for contrastive learning of visual representations. In: ICML. PMLR (2020)
[17]
Khosla, P., et al.: Supervised contrastive learning. In: NeurIPS, pp. 18661–18673 (2020)
[18]
Gunel, B., et al.: Supervised contrastive learning for pre-trained language model fine-tuning. In: ICLR (2020)
[19]
An, C., et al.: CoNT: contrastive neural text generation. In: NeurIPS (2022)
[20]
Salton G and Buckley C Term-weighting approaches in automatic text retrieval IPM 1988 24 5 513-523
[21]
Bahdanau, D., Cho, K.H., Bengio, Y.: Neural machine translation by jointly learning to align and translate. In: ICLR (2015)
[22]
Zhang Yu and Yang Q A survey on multi-task learning IEEE Trans. Knowl. Data Eng. 2021 34 12 5586-5609
[23]
Wolf, T., et al.: Transformers: State-of-the-art natural language processing. In: EMNLP: System Demonstrations (2020)
[24]
Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014)
[25]
Popović, M.: chrF++: words helping character N-grams. In: Proceedings of the second conference on machine translation (2017)
[26]
Lin, C.-Y.: ROUGE: a package for automatic evaluation of summaries. In: Text Summarization Branches Out (2004)
[27]
Vaswani, A., et al.: Attention is all you need. In: NeurIPS (2017)
Recommendations
Prediction, selection, and generation: a knowledge-driven conversation system
AbstractIn conversational systems, we can use external knowledge to generate more diverse sentences and make these sentences contain actual knowledge. Leveraging knowledge for conversation system is important but challenging. Firstly, the conversation ...
Knowledge-Grounded Dialogue with Reward-Driven Knowledge Selection
Natural Language Processing and Chinese ComputingAbstractKnowledge-grounded dialogue is a task of generating a fluent and informative response based on both conversation context and a collection of external knowledge, in which knowledge selection plays an important role and attracts more and more ...
Improving knowledge-based dialogue generation through two-stage knowledge selection and knowledge selection-guided pointer network
AbstractExisting End-to-End neural models for dialogue generation tend to generate generic and uninformative responses. Recently, knowledge-based dialogue models have been developed to generate more informative responses by leveraging external knowledge. ...
Comments
Information & Contributors
Information
Published In
Oct 2023
952 pages
ISBN:978-981-99-7253-1
DOI:10.1007/978-981-99-7254-8
- Editors:
- Feng Zhang,
- Hua Wang,
- Mahmoud Barhamgi,
- Lu Chen,
- Rui Zhou
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023.
Publisher
Springer-Verlag
Berlin, Heidelberg
Publication History
Published: 25 October 2023
Author Tags
Qualifiers
- Article
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 0Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 26 Sep 2024
Other Metrics
Citations
View Options
View options
Get Access
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in