WU Yunbing1,2, YE Chenglong1,2, YIN Aiying2,3, CHEN Kaizhi1,2, YANG Zhou1,2
1. College of Computer and Data Science, Fuzhou University, Fuzhou 350108 2. Digital Fujian Institute of Financial Big Data, Fuzhou University, Fuzhou 350108 3. Department of Computer Engineering, Zhicheng College, Fu-zhou University, Fuzhou 350002 on EmpatheticDialogues dataset indicate that PERG achieves superior performance in multiple metrics.
Abstract：Empathetic response generation aims to understand the experiences and feelings of users in conversations and provide appropriate responses. Psychological theories suggest that roles serve as an external manifestation of personality and are closely related to empathy. However, existing research primarily focuses on the cognitive and emotional factors of empathy while neglecting role factors that are beneficial to empathy, resulting in a lack of personalized empathetic responses. To address this issue, a persona-enhanced empathetic response generation model(PERG) is proposed. A persona-enhanced encoding module is introduced to capture deep semantic relationships among context, situation and role information through an encoder. By filtering role information based on context and situation, the understanding of the speaker′s and responder′s roles by the model is significantly improved, and thereby enhancing its empathetic capabilities. In the persona control decoding module, a multi-decoder control fusion mechanism is designed. The role information is effectively combined to regulate the impact of context and situation on empathy responses , generating highly personalized empathetic responses. Experiments
[1] 杨 州,陈志豪,蔡铁城,等.基于深度学习的情感对话响应综述.计算机学报, 2023, 46(12): 2489-2519. (YANG Z, CHEN Z H, CAI T C, et al. A Survey of Deep Learning Based Emotional Dialogue Response. Chinese Journal of Computers, 2023, 46(12): 2489-2519.) [2] LIANG Y L, MENG F D, ZHANG Y, et al. Infusing Multi-source Knowledge with Heterogeneous Graph Neural Network for Emotional Conversation Generation. Proceedings of the AAAI Conference on Artificial Intelligence, 2021, 35(15): 13343-13352. [3] LIU S Y, ZHENG C J, DEMASI O, et al. Towards Emotional Support Dialog Systems // Proc of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing(Long Papers). Stroudsburg, USA: ACL, 2021: 3469-3483. [4] RASHKIN H, SMITH E M, LI M, et al. Towards Empathetic Open-Domain Conversation Models: A New Benchmark and Dataset // Proc of the 57th Annual Meeting of the Association for Computational Linguistics. Stroudsburg, USA: ACL, 2019: 5370-5381. [5] XIE Y B, SVIKHNUSHINA E, PU P. A Multi-turn Emotionally Engaging Dialog Model[C/OL]. [2024-09-21]. https://arxiv.org/pdf/1908.07816v2. [6] 赵妍妍,陆 鑫,赵伟翔,等.情感对话技术综述.软件学报, 2024, 35(3): 1377-1402. (ZHAO Y Y, LU X, ZHAO W X, et al. Survey on Emotional Dialogue Techniques. Journal of Software, 2024, 35(3): 1377-1402.) [7] ZHENG C J, LIU Y, CHEN W, et al. CoMAE: A Multi-factor Hierarchical Framework for Empathetic Response Generation // Proc of the Joint Conference of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing. Stroudsburg, USA: ACL, 2021: 813-824. [8] DUDĂU D P, SAVA F A. Performing Multilingual Analysis with Linguistic Inquiry and Word Count 2015(LIWC2015). An Equivalence Study of Four Languages. Frontiers in Psychology, 2021, 12. DOI: 10.3389/fpsyg.2021.570568. [9] SHIN J, XU P, MADOTTO A, et al. Generating Empathetic Res-ponses by Looking Ahead the User′s Sentiment // Proc of the IEEE International Conference on Acoustics, Speech and Signal Proce-ssing. Washington, USA: IEEE, 2020: 7989-7993. [10] LIN Z J, MADOTTO A, SHIN J, et al. MoEL: Mixture of Empathetic Listeners // Proc of the Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Confe-rence on Natural Language Processing. Stroudsburg, USA: ACL, 2019: 121-132. [11] MAJUMDER N, HONG P F, PENG S S, et al. MIME: MIMi-cking Emotions for Empathetic Response Generation // Proc of the Conference on Empirical Methods in Natural Language Processing. Stroudsburg, USA: ACL, 2020: 8968-8979. [12] LI Q T, CHEN H S, REN Z C, et al. EmpDG: Multiresolution Interactive Empathetic Dialogue Generation // Proc of the 28th International Conference on Computational Linguistics. Stroudsburg, USA: ACL, 2020: 4454-4466. [13] ZHONG P X, WANG D, MIAO C Y. An Affect-Rich Neural Conversational Model with Biased Attention and Weighted Cross-Entropy Loss. Proceedings of the AAAI Conference on Artificial Intelligence, 2019, 33(1): 7492-7500. [14] ZHONG P X, WANG D, LI P F, et al. CARE: Commonsense-Aware Emotional Response Generation with Latent Concepts. Proceedings of the AAAI Conference on Artificial Intelligence, 2021, 35(16): 14577-14585. [15] LI Q T, LI P J, REN Z C, et al. Knowledge Bridging for Empathetic Dialogue Generation. Proceedings of the AAAI Conference on Artificial Intelligence, 2022, 36(10): 10993-11001. [16] SABOUR S, ZHENG C J, HUANG M L. CEM: Commonsense-Aware Empathetic Response Generation. Proceedings of the AAAI Conference on Artificial Intelligence, 2022, 36(10): 11229-11237. [17] WANG L R, LI J N, LIN Z, et al. Empathetic Dialogue Generation via Sensitive Emotion Recognition and Sensible Knowledge Selection // Proc of the Conference on Empirical Methods in Natural Language Processing. Stroudsburg, USA: ACL, 2022: 4634-4645. [18] ZHAO W X, ZHAO Y Y, LU X, et al. Don′t Lose Yourself! Empathetic Response Generation via Explicit Self-Other Awareness // Proc of the Findings of the Association for Computational Linguistics. Stroudsburg, USA: ACL, 2023: 13331-13344. [19] WANG Y T, CHEN C, YANG Z, et al. CTSM: Combining Trait and State Emotions for Empathetic Response Model // Proc of the Joint International Conference on Computational Linguistics, Language Resources and Evaluation. Stroudsburg, USA: ACL, 2024: 4214-4225. [20] YANG Z, REN Z C, WANG Y F, et al. An Iterative Associative Memory Model for Empathetic Response Generation // Proc of the 62nd Annual Meeting of the Association for Computational Linguistics. Stroudsburg, USA: ACL, 2024: 3081-3092. [21] YANG Z, REN Z C, WANG Y F, et al. Situation-Aware Empathetic Response Generation. Information Processing and Management, 2024, 61(6). DOI: 10.1016/j.ipm.2024.103824. [22] ZHOU J F, ZHENG C J, WANG B, et al. CASE: Aligning Coarse-to-Fine Cognition and Affection for Empathetic Response Generation // Proc of the 61st Annual Meeting of the Association for Computational Linguistics. Stroudsburg, USA: ACL, 2023: 8223-8237. [23] CAI H, SHEN X L, XU Q, et al. Improving Empathetic Dialogue Generation by Dynamically Infusing Commonsense Knowledge // Proc of the Findings of the Association for Computational Linguistics. Stroudsburg, USA: ACL, 2023: 7858-7873. [24] BI G Q, SHEN L, CAO Y N, et al. DIFFUSEMP: A Diffusion Model-Based Framework with Multi-grained Control for Empathetic Response Generation // Proc of the 61st Annual Meeting of the Association for Computational Linguistics(Long Papers). Stroudsburg, USA: ACL, 2023: 2812-2831. [25] RAAMKUMAR A S, YANG Y P. Empathetic Conversational Systems: A Review of Current Advances, Gaps, and Opportunities. IEEE Transactions on Affective Computing, 2023, 14(4): 2722-2739. [26] ZHONG P X, ZHANG C, WANG H, et al. Towards Persona-Based Empathetic Conversational Models // Proc of the Conference on Empirical Methods in Natural Language Processing. Stroudsburg, USA: ACL, 2020: 6556-6566. [27] GIORGI S, SEDOC J, BARRIERE V, et al. Findings of WASSA 2024 Shared Task on Empathy and Personality Detection in Interactions // Proc of the 14th Workshop on Computational Approaches to Subjectivity, Sentiment, and Social Media Analysis. Stroudsburg, USA: ACL, 2024: 369-379. [28] HOJAT M, MAIO V, POHL C A, et al. Clinical Empathy: Definition, Measurement, Correlates, Group Differences, Erosion, Enhancement, and Healthcare Outcomes. Discover Health Systems, 2023, 2(1). DOI: 10.1007/s44250-023-00020-2. [29] CHENG J L, SABOUR S, SUN H, et al. PAL: Persona-Augmented Emotional Support Conversation Generation // Proc of the Findings of the Association for Computational Linguistics. Stroudsburg, USA: ACL, 2023: 535-554. [30] BURKHARD S. The Relationship of Medical Provider Humility, Empathy, and Competency on Patient Satisfaction. Ph.D. Dissertation. Newberg, USA: George Fox University, 2023. [31] ZHANG S Z, DINAN E, URBANEK J, et al. Personalizing Dialogue Agents: I Have a Dog, Do You Have Pets Too? // Proc of the 56th Annual Meeting of the Association for Computational Linguistics(Long Papers). Stroudsburg, USA: ACL, 2018: 2204-2213. [32] SAP M, LE BRAS R, ALLAWAY E, et al. ATOMIC: An Atlas of Machine Commonsense for If-Then Reasoning. Proceedings of the AAAI Conference on Artificial Intelligence, 2019, 33(1): 3027-3035. [33] HWANG J D, BHAGAVATULA C, LE BRAS R, et al. (Comet-) Atomic 2020: On Symbolic and Neural Commonsense Knowledge Graphs. Proceedings of the AAAI Conference on Artificial Intelligence, 2021, 35(7): 6384-6392. [34] SEE A, LIU P J, MANNING C D. Get to the Point: Summarization with Pointer-Generator Networks // Proc of the 55th Annual Meeting of the Association for Computational Linguistics. Stroudsburg, USA: ACL, 2017: 1073-1083. [35] PENNINGTON J, SOCHER R, MANNING C D. GloVe: Global Vectors for Word Representation // Proc of the Conference on Empirical Methods in Natural Language Processing. Stroudsburg, USA: ACL, 2014: 1532-1543. [36] VASWANI A, SHAZEER N, PARMAR N, et al. Attention Is All You Need // Proc of the 31st International Conference on Neural Information Processing Systems. Cambridge, USA: MIT Press, 2017: 6000-6010. [37] TAHA K. Semi-Supervised and Unsupervised Clustering: A Review and Experimental Evaluation. Information Systems, 2023, 114. DOI: 10.1016/j.is.2023.102178. [38] YANG Z, REN Z C, WANG Y F, et al. Exploiting Emotion-Semantic Correlations for Empathetic Response Generation // Proc of the Conference on Empirical Methods in Natural Language Proce-ssing. Stroudsburg, USA: ACL, 2023: 4826-4837. [39] LI J W, GALLEY M, BROCKETT C, et al. A Diversity-Promoting Objective Function for Neural Conversation Models // Proc of the Conference of the North American Chapter of the Association for Computational Linguistics(Human Language Technologies). Strouds-burg, USA: ACL, 2016: 110-119. [40] SPEER R, CHIN J, HAVASI C. ConceptNet 5.5: An Open Multilingual Graph of General Knowledge. Proceedings of the AAAI Conference on Artificial Intelligence, 2017, 31(1): 4444-4451. [41] FLEISS J L, COHEN J. The Equivalence of Weighted Kappa and the Intraclass Correlation Coefficient as Measures of Reliability. Educational and Psychological Measurement, 1973, 33(3): 613-619.
2024, Vol. 37 
