1. Yunnan Key Laboratory of Intelligent Systems and Computing, Yunnan University, Kunming 650500; 2. School of Information Science and Engineering, Yunnan University, Kunming 650500
Abstract：There are two urgent challenges in conversational question answering to be addressed at present. One is how coreference and long range dependencies can be resolved to effectively utilize dependency information. The other is how contextual query subgraphs can be effectively maintained to avoid the risk of excessive expansion, thereby enabling more precise answer retrieval within them. In this paper, a model of conversational question answering based on knowledge graph and coreference resolution is proposed. First, coreference resolution is employed to obtain coreference clusters and an index replacement algorithm is introduced to enhance the semantic information of questions. Additionally, two types of dependency information, word coreference structure and character semantics, are proposed to guide the expansion of contextual query subgraph and answer retrieval. The contextual query subgraph is effectively expanded based on dependency information to obtain accurate query subgraph while avoiding overgrowth. Then, a reward-and-punishment mechanism is designed based on the number of dialogue rounds and the size of the query subgraph to effectively prevent the subgraph from overgrowing. Finally, dependency information is utilized to effectively improve the accuracy of answer retrieval. Experiments on the ConvQuestions dataset verify the effectiveness of the proposed method.
[1] ZAIB M, ZHANG W E, SHENG Q Z, et al. Conversational Question Answering: A Survey. Knowledge and Information Systems, 2022, 64(12): 3151-3195. [2] ZHONG L F, WU J, LI Q, et al. A Comprehensive Survey on Automatic Knowledge Graph Construction. ACM Computing Surveys, 2024, 56(4). DOI: 10.1145/361829. [3] 饶东宁,许正辉,梁瑞仕.基于知识库问答的回答生成研究.计算机工程, 2025, 51(2): 94-101. (RAO D N, XU Z H, LIANG R S.Research on Answer Generation Based on Knowledge Base Question Answering. Computer Enginee-ring, 2025, 51(2): 94-101.) [4] KACUPAJ E, SINGH K, MALESHKOVA M, et al. Contrastive Representation Learning for Conversational Question Answering over Knowledge Graphs // Proc of the 31st ACM International Confe-rence on Information and Knowledge Management. New York, USA: ACM, 2022: 925-934. [5] KAISER M, ROY R S, WEIKUM G.Reinforcement Learning from Reformulations in Conversational Question Answering over Know-ledge Graphs // Proc of the 44th International ACM SIGIR Confe-rence on Research and Development in Information Retrieval. New York, USA: ACM, 2021: 459-469. [6] KIM G, KIM H, PARK J, et al. Learn to Resolve Conversational Dependency: A Consistency Training Framework for Conversational Question Answering // 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: 6130-6141. [7] 陈晨,朱晴晴,严睿,等.基于深度学习的开放领域对话系统研究综述.计算机学报, 2019, 42(7): 1439-1466. (CHEN C, ZHU Q Q, YAN R, et al. Survey on Deep Learning Based Open Domain Dialogue System. Chinese Journal of Compu-ters, 2019, 42(7): 1439-1466.) [8] 宋鹏程,单丽莉,孙承杰,等.基于查询路径排序的知识库问答系统.中文信息学报, 2021, 35(11): 109-117, 126. (SONG P C, SHAN L L, SUN C J, et al. A Knowledge Base Question Answering System Based on Query Path Ranking. Journal of Chinese Information Processing, 2021, 35(11): 109-117, 126.) [9] 乔凯,陈可佳,陈景强.基于知识图谱与关键词注意机制的中文医疗问答匹配方法.模式识别与人工智能, 2021, 34(8): 733-741. (QIAO K, CHEN K J, CHEN J Q.Chinese Medical Question Answering Matching Method Based on Knowledge Graph and Keyword Attention Mechanism. Pattern Recognition and Artificial Intelligence, 2021, 34(8): 733-741.) [10] YIH W, CHANG M, HE X D, et al. Semantic Parsing via Staged Query Graph Generation: Question Answering with Knowledge Base // Proc of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Na-tural Language Processing(Long Papers). Stroudsburg, USA: ACL, 2015: 1321-1331. [11] SUN Y W, ZHANG L L, CHENG G, et al. SPARQA: Skeleton-Based Semantic Parsing for Complex Questions over Knowledge Bases. Proceedings of the AAAI Conference on Artificial Intelligence, 2020, 34(5): 8952-8959. [12] WANG S J, JIAO J, ZHANG X W.A Semantic Similarity-Based Subgraph Matching Method for Improving Question Answering over RDF // Proc of the ACM on Web Conference. New York, USA: ACM, 2020: 63-64. [13] CHRISTMANN P, ROY R S, ABUJABAL A, et al. Look Before You Hop: Conversational Question Answering over Knowledge Graphs Using Judicious Context Expansion // Proc of the 28th ACM International Conference on Information and Knowledge Ma-nagement. New York, USA: ACM, 2019: 729-738. [14] JIN J H, LUO J Z, KHEMMARAT S, et al. GStar: An Efficient Framework for Answering Top-k Star Queries on Billion-Node Knowledge Graphs. World Wide Web, 2019, 22(4): 1611-1638. [15] 成凌云,郭银章,刘青芳.基于对抗强化学习的多跳知识推理.模式识别与人工智能, 2025, 38(1): 22-35. (CHENG L Y, GUO Y Z, LIU Q F.Multi-hop Knowledge Reasoning Based on Adversarial Reinforcement Learning. Pattern Re-cognition and Artificial Intelligence, 2025, 38(1): 22-35.) [16] LAN Y S, JIANG J.Modeling Transitions of Focal Entities for Con-versational Knowledge Base Question Answering // 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: 3288-3297. [17] 李凤英,何晓蝶,董荣胜.融合语义信息的知识图谱多跳推理模型.模式识别与人工智能, 2022, 35(11): 1025-1032. (LI F Y, HE X D, DONG R S.Multi-hop Inference Model for Knowledge Graphs Incorporating Semantic Information. Pattern Recognition and Artificial Intelligence, 2022, 35(11): 1025-1032.) [18] REN H Y, HU W H, LESKOVEC J.Query2box: Reasoning over Knowledge Graphs in Vector Space Using Box Embeddings[C/OL]. [2025-10-16]. https://arxiv.org/pdf/2002.05969. [19] QU C, YANG L, QIU M H, et al. Attentive History Selection for Conversational Question Answering // Proc of the 28th ACM International Conference on Information and Knowledge Management. New York, USA: ACM, 2019: 1391-1400. [20] QIU M H, HUANG X J, CHEN C, et al. Reinforced History Back-tracking for Conversational Question Answering. Proceedings of the AAAI Conference on Artificial Intelligence, 2021, 35(15): 13718-13726. [21] LIU L H, HILL B, DU B X, et al. Conversational Question Answering with Language Models Generated Reformulations over Knowledge Graph // Findings of the Association for Computational Linguistics. Stroudsburg, USA: ACL, 2024: 839-850. [22] KAISER M, ROY R S, WEIKUM G.Robust Training for Conversational Question Answering Models with Reinforced Reformulation Generation // Proc of the 17th ACM International Conference on Web Search and Data Mining. New York, USA: ACM, 2024: 322-331. [23] KE X R, ZHANG J, LÜ X, et al. Knowledge-Augmented Self-Training of a Question Rewriter for Conversational Knowledge Base Question Answering // Findings of the Association for Computational Linguistics. Stroudsburg, USA: ACL, 2022: 1844-1856. [24] SU H, SHEN X Y, ZHANG R Z, et al. Improving Multi-turn Dialogue Modelling with Utterance Rewriter // Proc of the 57th Annual Meeting of the Association for Computational Linguistics. Stroudsburg, USA: ACL, 2019: 22-31. [25] QUAN J, XIONG D Y, WEBBER B, et al. GECOR: An End-to-End Generative Ellipsis and Coreference Resolution Model for Task-Oriented Dialogue // Proc of the Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing. Stroudsburg, USA: ACL, 2019: 4546-4556. [26] 张诗安,熊德意.使用共指消解增强多轮任务型对话生成.中文信息学报, 2022, 36(9): 149-158. (ZHANG S A, XIONG D Y.Improving Multi-turn Task-Oriented Dialogue Generation Using Coreference Resolution. Journal of Chinese Information Processing, 2022, 36(9): 149-158.) [27] SUN J S, XU C J, TANG L M Y, et al. Think-on-Graph: Deep and Responsible Reasoning of Large Language Model on Knowledge Graph[C/OL].[2025-10-16]. https://arxiv.org/pdf/2307.07697. [28] KAISER M, WEIKUM G.Preference-Based Learning with Retrie-val Augmented Generation for Conversational Question Answering // Proc of the ACM on Web Conference. New York, USA: ACM, 2025: 1053-1057. [29] AGARWAL P, BEDATHUR S.A Zero-Shot Neuro-Symbolic Approach for Complex Knowledge Graph Question Answering // Findings of the Association for Computational Linguistics. Stroudsburg, USA: ACL, 2025: 11514-11527. [30] LEE K, HE L H, LEWIS M, et al. End-to-End Neural Corefe-rence Resolution // Proc of the Conference on Empirical Methods in Natural Language Processing. Stroudsburg, USA: ACL, 2017: 188-197. [31] FERRAGINA P, SCAIELLA U.TAGME: On-the-Fly Annotation of Short Text Fragments(by Wikipedia Entities) // Proc of the 19th ACM International Conference on Information and Knowledge Management. New York, USA: ACM, 2010: 1625-1628. [32] MIKOLOV T, SUTSKEVER I, CHEN K, et al. Distributed Representations of Words and Phrases and Their Compositionality // Proc of the 27th International Conference on Neural Information Proce-ssing Systems. Cambridge, USA: MIT Press, 2013: 3111-3119. [33] FAGIN R, LOTEM A, NAOR M.Optimal Aggregation Algorithms for Middleware. Journal of Computer and System Sciences, 2003, 66(4): 614-656. [34] GUO D Y, TANG D Y, DUAN N, et al. Dialog-to-Action: Conversational Question Answering over a Large-Scale Knowledge Base // Proc of the 32nd International Conference on Neural Information Processing Systems. Cambridge, USA: MIT Press, 2018: 2946-2955. [35] LI Y Q, YANG N, WANG L, et al. Generative Retrieval for Conversational Question Answering. Information Processing and Ma-nagement, 2023, 60(5). DOI: 10.1016/j.ipm.2023.103475. [36] CHRISTMANN P, ROY R S, WEIKUM G.Explainable Conversational Question Answering over Heterogeneous Sources via Iterative Graph Neural Networks // Proc of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York, USA: ACM, 2023: 643-653. [37] LIU Z Q, GAN C T, WANG J J, et al. OntoTune: Ontology-Driven Self-Training for Aligning Large Language Models // Proc of the ACM on Web Conference. New York, USA: ACM, 2025: 119-133. [38] DEVLIN J, CHANG M W, LEE K, et al. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding // Proc of the Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies(Long and Short Papers). Stroudsburg, USA: ACL, 2019: 4171-4186. [39] JOSHI M, CHEN D Q, LIU Y H, et al. Spanbert: Improving Pre-Training by Representing and Predicting Spans. Transactions of the Association for Computational Linguistics, 2020, 8: 64-77.
2026, Vol. 39 
