Persona Consistency Response Generation by Integrating Dialogue Coherence Memory

doi:10.3778/j.issn.1002-8331.2306-0274

Abstract

Abstract: The existing research on dialogue systems often overlooks the issue of persona consistency during the dialogue process, which may lead to persona mismatches before and after the dialogue, thereby reducing the user’s dialogue experience. At the same time, the issue of coherence before and after the dialogue process is also of great concern. To address the above issues, a persona consistency response generation method that integrates dialogue coherence memory is proposed. Firstly, the original dataset is persona-extended through the persona extension module to obtain the persona-extended dataset. Secondly, the highly correlated persona information and dialogue history are extracted through the data extraction module, and redundant information is removed to obtain the training dataset. Finally, the persona consistency memory module and dialogue coherence memory module are introduced in the response generation module, so that the dialogue model always maintains the memory characteristics of persona information and dialogue history during the dialogue process, promoting its generation of dialogue-coherent persona consistent response. The experimental results on the Persona-Chat dataset show that the responses generated by this method have been improved in terms of dialogue coherence and persona consistency.

Key words: persona consistency response, persona extension, data extraction, persona consistency memory, dialogue coherence memory

摘要： 现有的对话系统研究往往忽略了对话过程中的角色一致性问题，这可能会导致对话前后的角色不匹配现象，从而降低用户的对话体验，同时，对话过程中的前后连贯性问题也备受人们所关注。为了解决以上问题，提出了一种融合对话连贯记忆的角色一致性回复生成方法。通过角色扩展模块将原始数据集进行角色扩展，得到角色扩展数据集；通过数据提取模块将相关度较高的角色信息与对话历史进行提取，并从中去除冗余信息，得到训练数据集；在回复生成模块中引入角色一致记忆模块和对话连贯记忆模块，使得对话模型在对话过程中始终保持对角色信息以及对话历史的记忆特性，促使其生成对话连贯的角色一致性回复。在Persona-Chat数据集上的实验结果表明，该方法生成的回复在对话连贯性以及角色一致性方面都得到了一定的提升。

关键词: 角色一致性回复生成, 角色扩展, 数据提取, 角色一致记忆, 对话连贯记忆

ZHU Chuanrun, TANG Hong, WANG Ningzhe, LIU Zhong. Persona Consistency Response Generation by Integrating Dialogue Coherence Memory[J]. Computer Engineering and Applications, 2024, 60(18): 167-175.

朱传润, 唐宏, 王宁喆, 刘钟. 融合对话连贯记忆的角色一致性回复生成方法[J]. 计算机工程与应用, 2024, 60(18): 167-175.

References

[1] XU X, GOU Z, WU W, et al. Long time no see! Open-domain conversation with long-term persona memory[C]//Proceedings of the Findings of the Association for Computational Linguistics (ACL 2022), 2022: 2639-2650.
[2] SONG H, WANG Y, ZHANG W, et al. Generate, delete and rewrite: a three-stage framework for improving persona consistency of dialogue generation[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020: 5821-5831.
[3] NI J, YOUNG T, PANDELEA V, et al. Recent advances in deep learning based dialogue systems: a systematic survey[J]. Artificial Intelligence Review, 2023, 56(4): 3055-3155.
[4] ZHANG S, DINAN E, URBANEK J, et al. Personalizing dialogue agents: I have a dog, do you have pets too?[C]//Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 2018: 2204-2213.
[5] WELLECK S, WESTON J, SZLAM A, et al. Dialogue natural language inference[C]//Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, 2019: 3731-3741.
[6] LIU Q, CHEN Y, CHEN B, et al. You impress me: dialogue generation via mutual persona perception[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020: 1417-1427.
[7] SONG H, WANG Y, ZHANG K, et al. BoB: BERT over BERT for training persona-based dialogue models from limited personalized data[C]//Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), 2021: 167-177.
[8] CAO Y, BI W, FANG M, et al. A model-agnostic data manipulation method for persona-based dialogue generation[C]//Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 2022: 7984-8002.
[9] STENT A, MARGE M, SINGHAI M. Evaluating evaluation methods for generation in the presence of variation[C]//International Conference on Intelligent Text Processing and Computational Linguistics, 2005: 341-351.
[10] CAMPOS J, KENNEDY J, LEHMAN J F. Challenges in exploiting conversational memory in human-agent interaction[C]//Proceedings of the 17th International Conference on Autonomous Agents and Multiagent System, 2018: 1649-1657.
[11] LI C, GAO X, LI Y, et al. Optimus: organizing sentences via pre-trained modeling of a latent space[C]//Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2020: 4678-4699.
[12] BAO S, HE H, WANG F, et al. PLATO: pre-trained dialogue generation model with discrete latent variable[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020: 85-96.
[13] CHEN W, GONG Y, WANG S, et al. DialogVED: a pre-trained latent variable encoder-decoder model for dialog response generation[C]//Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 2022: 4852-4864.
[14] XIE Q, DAI Z, HOVY E, et al. Unsupervised data augmentation for consistency training[C]//Advances in Neural Information Processing Systems, 2020: 6256-6268.
[15] SAP M, LE BRAS R, ALLAWAY E, et al. Atomic: an atlas of machine commonsense for if-then reasoning[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2019: 3027-3035.
[16] BOSSELUT A, RASHKIN H, SAP M, et al. COMET: commonsense transformers for automatic knowledge graph construction[C]//Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, 2019: 4762-4779.
[17] ZHAO T, ZHAO R, ESKENAZI M. Learning discourse-level diversity for neural dialog models using conditional variational autoencoders[C]//Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, 2017: 654-664.
[18] WOLF T, SANH V, CHAUMOND J, et al. Transfertransfo: a transfer learning approach for neural network based conversational agents[J]. arXiv:1901.08149, 2019.
[19] LEWIS M, LIU Y, GOYAL N, et al. BART: denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020: 7871-7880.
[20] LOSHCHILOV I, HUTTER F. Decoupled weight decay regularization[J]. arXiv:1711.05101, 2017.
[21] SONG H, ZHANG W N, CUI Y, et al. Exploiting persona information for diverse generation of conversational responses[J]. arXiv:1905.12188, 2019.
[22] LIAN R, XIE M, WANG F, et al. Learning to select knowledge for response generation in dialog systems[C]//Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, 2019: 5081.
[23] BAHL L R, JELINEK F, MERCER R L. A maximum likelihood approach to continuous speech recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1983(2): 179-190.
[24] PAPINENI K, ROUKOS S, WARD T, et al. Bleu: a method for automatic evaluation of machine translation[C]//Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, 2002: 311-318.
[25] LI J, GALLEY M, BROCKETT C, et al. A diversity-promoting objective function for neural conversation models[C]//Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, 2016: 110-119.
[26] MADOTTO A, LIN Z, WU C S, et al. Personalizing dialogue agents via meta-learning[C]//Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, 2019: 5454-5459.
[27] ZUEHLKE D, GEWENIGER T, HEIMANN U, et al. Fuzzy fleiss-kappa for comparison of fuzzy classifiers[C]//Proceedinds of 17th European Symposium on Artificial Neural Networks, Bruges, Belgium, April 22-24, 2009.