知识驱动的对话生成模型研究综述

doi:10.3778/j.issn.1673-9418.2304022

摘要/Abstract

摘要： 知识驱动的对话生成模型旨在利用不同形式的知识来强化对话生成模型，使得对话生成模型不仅能从对话数据中学习语义交互，而且还能深入理解用户输入、背景知识和对话上下文，生成更合理、更具多样性、更富含信息量和拟人的回复，进而推动对话系统的发展。目前相关工作仍处于初期探索阶段，并且很少有对现有成果的全面梳理和系统总结。对知识驱动的对话生成模型研究展开综述，首先，针对现有的研究成果，梳理并介绍了当前知识驱动的对话生成任务和主要遇到的问题，并且给出详细的任务定义和问题定义；其次，整理并介绍了知识驱动的对话生成模型建模所需的数据集；然后，对目前知识驱动的对话生成研究过程中知识获取、知识表示、知识选择和知识融入相关研究中每个模型的改进、研究现状、模型涉及的评价指标和模型的性能进行重点介绍；最后，对知识驱动的对话生成模型研究未来的发展方向进行展望。

关键词: 对话生成模型, 对话系统, 外部知识, 知识驱动的对话生成

Abstract: Knowledge-driven dialogue generation models aim to enhance dialogue generation models by using different forms of knowledge, so that dialogue generation models can not only learn semantic interactions from dialogue data, but also deeply understand user input, background knowledge and dialogue context to generate more reasonable, diverse, informative and anthropomorphic responses, and thus promote the development of dialogue systems. Currently, the related work is still in the early stages of exploration, and there is a lack of comprehensive reviews and systematic summaries of existing results. This paper provides a comprehensive review of the research on knowledge-driven dialogue generation models. Firstly, in response to the existing research results, it sorts out and introduces the current knowledge-driven dialogue generation tasks and the main problems encountered, and provides detailed task definitions and problem definitions. Secondly, it organizes and introduces the datasets required for the modeling of knowledge-driven dialogue generation models. Then, it focuses on the improvement, research status, evaluation indicators involved, and performance of each model in the process of knowledge-driven dialogue generation research, including knowledge acquisition, knowledge representation, knowledge selection, and knowledge integration-related studies. Finally, the future development directions of knowledge-based dialogue generation models are prospected.

Key words: dialogue generation model, dialogue systems, external knowledge, knowledge-driven dialogue generation

许璧麒, 马志强, 周钰童, 贾文超, 刘佳, 吕凯. 知识驱动的对话生成模型研究综述[J]. 计算机科学与探索, 2024, 18(1): 58-74.

XU Biqi, MA Zhiqiang, ZHOU Yutong, JIA Wenchao, LIU Jia, LYU Kai. Survey of Research on Knowledge-Driven Dialogue Generation Models[J]. Journal of Frontiers of Computer Science and Technology, 2024, 18(1): 58-74.

参考文献

[1] TURING A M. Computing machinery and intelligence[J]. Mind, 1950, 59(236): 433-460.
[2] ARON J. How innovative is apple??s new voice assistant, Siri?[J]. New Scientist, 2011, 212(2836): 24.
[3] QIU M, LI F L, WANG S, et al. Alime chat: a sequence to sequence and rerank based chatbot engine[C]//Proceedings of the 55th Annual Meeting of the Association for Com-putational Linguistics. Stroudsburg: ACL, 2017: 498-503.
[4] ZHOU L, GAO J, LI D, et al. The design and implementa-tion of xiaoice, an empathetic social chatbot[J]. Computat-ional Linguistics, 2020, 46(1): 53-93.
[5] WEIZENBAUM J. ELIZA—a computer program for the study of natural language communication between man and ma-chine[J]. Communications of the ACM, 1983, 26(1): 23-28.
[6] SUTSKEVER I, VINYALS O, LE Q V. Sequence to sequence learning with neural networks[C]//Advances in Neural In-formation Processing Systems 27, Montreal, Dec 8-13, 2014: 3104-3112.
[7] LOWE R, POW N, SERBAN I, et al. Incorporating unstruc-tured textual knowledge sources into neural dialogue sys-tems[C]//Neural Information Processing Systems Work-shop on Machine Learning for Spoken Language Unders-tanding, Montreal, Dec 7-12, 2015.
[8] DODGE J, GANE A, ZHANG X, et al. Evaluating prere-quisite qualities for learning end-to-end dialog systems[J]. arXiv:1511.06931, 2015.
[9] MAHDISOLTANI F, BIEGA J, SUCHANEK F. Yago3: a knowledge base from multilingual wikipedias[C]//Procee-dings of the 7th Biennial Conference on Innovative Data Systems Research, Asilomar, Jan 4-7, 2015.
[10] VRANDE?I? D, KR?TZSCH M. Wikidata: a free collabo-rative knowledgebase[J]. Communications of the ACM, 2014, 57(10): 78-85.
[11] BOLLACKER K, COOK R, TUFTS P. Freebase: a shared database of structured general human knowledge[C]//Pro-ceedings of the 22nd AAAI Conference on Artificial Intel-ligence, Vancouver, Jul 22-26, 2007: 1962-1963.
[12] BIZER C, LEHMANN J, KOBILAROV G, et al. DBpedia—a crystallization point for the web of data[J]. Journal of Web Semantics, 2009, 7(3): 154-165.
[13] SPEER R, HAVASI C. ConceptNet 5: a large semantic net-work for relational knowledge[M]//The People’s Web Meets NLP: Collaboratively Constructed Language Resources.Berlin, Heidelberg: Springer, 2013: 161-176.
[14] SWARTZ A. Musicbrainz: a semantic web service[J]. IEEE Intelligent Systems, 2002, 17(1): 76-77.
[15] AHLERS D. Assessment of the accuracy of GeoNames gazetteer data[C]//Proceedings of the 7th Workshop on Geo-graphic Information Retrieval, Beijing, May 15-16, 2013: 74-81.
[16] WISHART D S, KNOX C, GUO A C, et al. DrugBank: a comprehensive resource for in silico drug discovery and exploration[J]. Nucleic Acids Research, 2006, 34: D668-D672.
[17] WANG X, GAO T, ZHU Z, et al. KEPLER: a unified model for knowledge embedding and pre-trained language repre-sentation[J]. Transactions of the Association for Computat-ional Linguistics, 2021, 9: 176-194.
[18] HAO J, CHEN M, YU W, et al. Universal representation learning of knowledge bases by jointly embedding ins-tances and ontological concepts[C]//Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Anchorage, Aug 4-8, 2019: 1709-1719.
[19] CAMBRIA E, SONG Y, WANG H, et al. Semantic multi-dimensional scaling for open-domain sentiment analysis[J]. IEEE Intelligent Systems, 2012, 29(2): 44-51.
[20] XU Z, LIU B, WANG B, et al. Incorporating loose-struc-tured knowledge into conversation modeling via recall-gate LSTM[C]//Proceedings of the 2017 International Joint Con-ference on Neural Networks, Alaska, May 14-18, 2017: 3506-3513.
[21] ZHU W, MO K, ZHANG Y, et al. Flexible end-to-end dialo-gue system for knowledge grounded conversation[J]. arXiv:1709.04264, 2017.
[22] ZHOU H, YOUNG T, HUANG M, et al. Commonsense knowledge aware conversation generation with graph atten-tion[C]//Proceedings of the 27th International Joint Con-ference on Artificial Intelligence, Stockholm, Jul 13-19, 2018: 4623-4629.
[23] LIU Z, NIU Z Y, WU H, et al. Knowledge aware conversa-tion generation with explainable reasoning over augmented graphs[J]. arXiv:1903.10245, 2019.
[24] ZHOU K, PRABHUMOYE S, BLACK A W. A dataset for document grounded conversations[J]. arXiv:1809.07358, 2018.
[25] LI Z, NIU C, MENG F, et al. Incremental transformer with deliberation decoder for document grounded conversations[J]. arXiv:1907.08854, 2019.
[26] FENG S, WAN H, GUNASEKARA C, et al. doc2dial: a goal-oriented document-grounded dialogue dataset[J]. arXiv:2011.06623, 2020.
[27] DINAN E, ROLLER S, SHUSTER K, et al. Wizard of Wikipedia: knowledge-powered conversational agents[J]. arXiv:1811.01241, 2018.
[28] ZHOU K, PRABHUMOYE S, BLACK A W. A dataset for document grounded conversations[J]. arXiv:1809.07358, 2018.
[29] MOGHE N, ARORA S, BANERJEE S, et al. Towards exploiting background knowledge for building conversa-tion systems[J]. arXiv:1809.08205, 2018.
[30] MAJUMDER B P, LI S, NI J, et al. Interview: large-scale modeling of media dialog with discourse patterns and know-ledge grounding[C]//Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, Punta Cana, Nov 11-12, 2020: 8129-8141.
[31] RODRIGUEZ P, CROOK P, MOON S, et al. Information seeking in the spirit of learning: a dataset for conversational curiosity[J]. arXiv:2005.00172, 2020.
[32] ZHOU H, ZHENG C, HUANG K, et al. KdConv: a Chinese multi-domain dialogue dataset towards multi-turn know-ledge-driven conversation[J]. arXiv:2004.04100, 2020.
[33] FAN A, JERNITE Y, PEREZ E, et al. ELI5: long form question answering[J]. arXiv:1907.09190, 2019.
[34] SHANG L F, LU Z D, LI H. Neural responding machine for short-text conversation[C]//Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing. Stroudsburg: ACL, 2015: 1577-1586.
[35] ERIC M, CHARTIER N, HEDAYATNIA B, et al. Multi-sentence knowledge selection in open-domain dialogue[C]//Proceedings of the 14th International Conference on Natural Language Generation, Aberdeen, Sep 20-24, 2021: 76-86.
[36] ZHANG H, LIU Z, XIONG C, et al. Grounded conversa-tion generation as guided traverses in commonsense know-ledge graphs[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. Strouds-burg: ACL, 2020: 2031-2043.
[37] GOPALAKRISHNAN K, HEDAYATNIA B, CHEN Q, et al. Topical-chat: towards knowledge-grounded open-domain conversations[C]//Proceedings of the 20th Annual Conference of the International Speech Communication Association, Graz, Sep 15-19, 2019: 1891-1895.
[38] RADFORD A, WU J, CHILD R, et al. Language models are unsupervised multitask learners[J]. OpenAI Blog, 2019, 1(8): 9.
[39] BROWN T, MANN B, RYDER N, et al. Language models are few-shot learners[C]//Advances in Neural Information Processing Systems 33, Dec 6-12, 2020: 1877-1901.
[40] OUYANG L, WU J, JIANG X, et al. Training language models to follow instructions with human feedback[C]//Advances in Neural Information Processing Systems 35, 2022: 27730-27744.
[41] SCAO T L, FAN A, AKIKI C, et al. Bloom: a 176b-parameter open-access multilingual language model[J]. arXiv:2211.05100, 2022.
[42] DEVLIN J, CHANG M W, LEE K, et al. BERT: pre-training of deep bidirectional transformers for language under-standing[C]//Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Minneapolis, Jun 2-7, 2019: 4171-4186.
[43] LIU Y, OTT M, GOYAL N, et al. RoBERTa: a robustly op-timized BERT pretraining approach[J]. arXiv:1907.11692,2019.
[44] RAFFEL C, SHAZEER N, ROBERTS A, et al. Exploring the limits of transfer learning with a unified text-to-text transformer[J]. The Journal of Machine Learning Research, 2020, 21(1): 5485-5551.
[45] SONG K, TAN X, QIN T, et al. MASS: masked sequence to sequence pre-training for language generation[C]//Pro-ceedings of the 2019 International Conference on Machine Learning, Long Beach, Jun 10-15, 2019: 5926-5936.
[46] 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 Lin-guistics, Seattle, Jul 5-10, 2020: 7871-7880.
[47] ACHILLE A, ROVERE M, SOATTO S. Critical learning periods in deep neural networks[J]. arXiv:1711.08856, 2017.
[48] LIU L Z, WANG Y, KASAI J, et al. Probing across time: what does RoBERTa know and when?[J]. arXiv:2104.07885, 2021.
[49] SAPHRA N, LOPEZ A. Understanding learning dynamics of language models with SVCCA[C]//Proceedings of the 2019 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Florence, Jul 28-31, 2019: 3257-3267.
[50] SAPHRA N, LOPEZ A. LSTMs compose-and learn-bottom-up[C]//Findings of the Association for Computational Linguistics: EMNLP 2020, Hong Kong, China, Nov 16-20, 2020: 2797-2809.
[51] GRAVES A, GRAVES A. Long short-term memory[J]. Neural Computation, 2010, 9(8): 1735-1780.
[52] RAGHU M, GILMER J, YOSINSKI J, et al. SVCCA: singular vector canonical correlation analysis for deep learning dynamics and interpretability[C]//Advances in Neural Infor-mation Processing Systems 30, Long Beach, Dec 4-9, 2017: 6076-6085.
[53] CHIANG C H, HUANG S F, LEE H Y. Pretrained language model embryology: the birth of ALBERT[C]//Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, Hong Kong, China, Nov 16-20, 2020: 6813-6828.
[54] LAN Z, CHEN M, GOODMAN S, et al. ALBERT: a lite BERT for self-supervised learning of language represen-tations[J]. arXiv:1909.11942, 2019.
[55] PéREZ-MAYOS L, BALLESTEROS M, WANNER L. How much pretraining data do language models need to learn syntax?[C]//Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, Putnta Cana, Nov 7-11, 2021: 1571-1582.
[56] LIU Y, OTT M, GOYAL N, et al. RoBERTa: a robustly opti-mized BERT pretraining approach[J]. arXiv:1907.11692, 2019.
[57] LIU Z, WANG Y, KASAI J, et al. Probing across time: what does RoBERTa know and when?[C]//Findings of the Association for Computational Linguistics: EMNLP 2021, Punta Cana, Nov 7-11, 2021: 820-842.
[58] SUN Y, WANG S, LI Y, et al. Ernie: enhanced represen-tation through knowledge integration[J]. arXiv:1904.09223, 2019.
[59] SHEN T, MAO Y, HE P, et al. Exploiting structured know-ledge in text via graph-guided representation learning[C]//Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, Hong Kong, China, Nov 16-20, 2020: 8980-8994.
[60] XIONG W, DU J, WANG W Y, et al. Pretrained encyclope-dia: weakly supervised knowledge-pretrained language model[J]. arXiv:1912.09637, 2019.
[61] YAMADA I, ASAI A, SHINDO H, et al. LUKE: deep con-textualized entity representations with entity-aware self-attention[C]//Proceedings of the 2020 Conference on Empi-rical Methods in Natural Language Processing, Hong Kong, China, Nov 16-20, 2020: 6442-6454.
[62] FéVRY T, SOARES L B, FITZGERALD N, et al. Entities as experts: sparse memory access with entity supervision[C]//Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, Hong Kong, China, Nov 16-20, 2020: 4937-4951.
[63] LOGESWARAN L, CHANG M W, LEE K, et al. Zero-shot entity linking by reading entity descriptions[C]//Procee-dings of the 57th Annual Meeting of the Association for Computational Linguistics, Florence, Jul 28-31, 2019: 3449-3460.
[64] GILLICK D, KULKARNI S, LANSING L, et al. Learning dense representations for entity retrieval[C]//Proceedings of the 23rd Conference on Computational Natural Language Learning, Hong Kong, China, Nov 3-4, 2019: 528-537.
[65] PETERS M E, NEUMANN M, LOGAN R, et al. Know-ledge enhanced contextual word representations[C]//Procee-dings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, Hong Kong, China, Nov 3-7, 2019: 43-54.
[66] ZHANG Z, HAN X, LIU Z, et al. ERNIE: enhanced lan-guage representation with informative entities[C]//Procee-dings of the 57th Annual Meeting of the Association for Computational Linguistics, Florence, Jul 28-31, 2019: 1441-1451.
[67] WANG X, GAO T, ZHU Z, et al. KEPLER: a unified model for knowledge embedding and pre-trained language repre-sentation[J]. Transactions of the Association for Computa-tional Linguistics, 2021, 9: 176-194.
[68] WANG R, TANG D, DUAN N, et al. K-Adapter: infusing knowledge into pre-trained models with adapters[C]//Fin-dings of the Association for Computational Linguistics: ACL-IJCNLP 2021, Bangkok, Aug 7, 2021: 1405-1418.
[69] QIN Y, LIN Y, TAKANOBU R, et al. ERICA: improving entity and relation understanding for pre-trained language models via contrastive learning[C]//Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, Bangkok, Aug 7, 2021: 3350-3363.
[70] LIU W, ZHOU P, ZHAO Z, et al. K-BERT: enabling language representation with knowledge graph[C]//Proceedings of the 2020 AAAI Conference on Artificial Intelligence, New York, Feb 7-12, 2020: 2901-2908.
[71] SOARES L B, FITZGERALD N, LING J, et al. Matching the blanks: distributional similarity for relation learning[C]//Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, Florence, Jul 28-31, 2019: 2895-2905.
[72] BOSSELUT A, RASHKIN H, SAP M, et al. COMET: com-monsense transformers for automatic knowledge graph construction[C]//Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, Florence, Jul 28-31, 2019: 4762-4779.
[73] GUAN J, HUANG F, ZHAO Z, et al. A knowledge-enhanced pretraining model for commonsense story genera-tion[J]. Transactions of the Association for Computational Linguistics, 2020, 8: 93-108.
[74] SHWARTZ V, WEST P, LE BRAS R, et al. Unsupervised commonsense question answering with self-talk[C]//Procee-dings of the 2020 Conference on Empirical Methods in Natural Language Processing, Hong Kong, China, Nov 16-20, 2020: 4615-4629.
[75] MA K, ILIEVSKI F, FRANCIS J, et al. Knowledge-driven data construction for zero-shot evaluation in commonsense question answering[C]//Proceedings of the 2021 AAAI Con-ference on Artificial Intelligence. Menlo Park: AAAI, 2021:13507-13515.
[76] KE P, JI H, LIU S, et al. SentiLARE: sentiment-aware language representation learning with linguistic knowledge[C]//Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, Hong Kong, China, Nov 16-20, 2020: 6975-6988.
[77] TIAN H, GAO C, XIAO X, et al. SKEP: sentiment know-ledge enhanced pre-training for sentiment analysis[C]//Pro-ceedings of the 58th Annual Meeting of the Association for Computational Linguistics, Seattle, Jul 5-10, 2020: 4067-4076.
[78] LAUSCHER A, VULI? I, PONTI E M, et al. Specializing unsupervised pretraining models for word-level semantic similarity[C]//Proceedings of the 28th International Confe-rence on Computational Linguistics, Barcelona, Dec 8-13, 2020: 1371-1383.
[79] MILLER G A. WordNet: a lexical database for English[J]. Communications of the ACM, 1995, 38(11): 39-41.
[80] NAVIGLI R, PONZETTO S P. BabelNet: building a very large multilingual semantic network[C]//Proceedings of the 48th Annual Meeting of the Association for Computational Linguistics, Uppsala, Jul 11-16, 2010: 216-225.
[81] SONG J, LIANG D, LI R, et al. Improving semantic mat-ching through dependency-enhanced pre-trained model with adaptive fusion[J]. arXiv:2210.08471, 2022.
[82] SACHAN D, ZHANG Y, QI P, et al. Do syntax trees help pre-trained transformers extract information?[C]//Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics, Apr 19-23, 2021: 2647-2661.
[83] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Advances in Neural Information Processing Systems 30, Long Beach, Dec 4-9, 2017: 5998-6008.
[84] BAI J, WANG Y, CHEN Y, et al. Syntax-BERT: improving pre-trained transformers with syntax trees[C]//Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics, Apr 19-23, 2021: 3011-3020.
[85] DAI D, DONG L, HAO Y, et al. Knowledge neurons in pre-trained transformers[C]//Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics, Dublin, May 22-27, 2022: 8493-8502.
[86] GEVA M, SCHUSTER R, BERANT J, et al. Transformer feed-forward layers are key-value memories[C]//Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, Punta Cana, Nov 7-11, 2021: 5484-5495.
[87] MENG K, BAU D, ANDONIAN A, et al. Locating and edi-ting factual knowledge in GPT[J]. arXiv:2202.05262, 2022.
[88] CLARK K, KHANDELWAL U, LEVY O, et al. What does BERT look at? An analysis of BERT’s attention[C]//Pro-ceedings of the 2019 ACL Workshop BlackboxNLP: Analy-zing and Interpreting Neural Networks for NLP, Florence, Jul 28-31, 2019: 276-286.
[89] HTUT P M, PHANG J, BORDIA S, et al. Do attention heads in BERT track syntactic dependencies?[J]. arXiv:1911.12246, 2019.
[90] LIN Y, TAN Y C, FRANK R. Open sesame: getting inside BERT’s linguistic knowledge[C]//Proceedings of the 2019 ACL Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP, Florence, Jul 28-31, 2019: 241-253.
[91] LIU N F, GARDNER M, BELINKOV Y, et al. Linguistic knowledge and transferability of contextual representations[C]//Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Minneapolis, Jun 2-7, 2019: 1073-1094.
[92] WALLAT J, SINGH J, ANAND A. BERTnesia: inves-tigating the capture and forgetting of knowledge in BERT[C]//Proceedings of the 3rd BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP. Stroud-sburg: ACL, 2020: 174-183.
[93] PETRONI F, ROCKT?SCHEL T, RIEDEL S, et al. Language models as knowledge bases?[C]//Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, Hong Kong, China, Nov 3-7, 2019: 2463-2473.
[94] JUNEJA J, AGARWAL R. Finding patterns in knowledge attribution for transformers[J]. arXiv:2205.01366, 2022.
[95] LIAN R, XIE M, WANG F, et al. Learning to select know-ledge for response generation in dialog systems[C]//Procee-dings of the 28th International Joint Conference on Arti-ficial Intelligence, Macao, Aug 10-16, 2019: 5081-5087.
[96] ZHANG Y, REN P, DE RIJKE M. Improving background based conversation with context-aware knowledge pre-selec-tion[J]. arXiv:1906.06685, 2019.
[97] DINAN E, ROLLER S, SHUSTER K, et al. Wizard of Wiki-pedia: knowledge-powered conversational agents[J]. arXiv:1811.01241, 2018.
[98] LIN X, JIAN W, HE J, et al. Generating informative conver-sational response using recurrent knowledge-interaction and knowledge-copy[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, Seattle, Jul 5-10, 2020: 41-52.
[99] KIM B, AHN J, KIM G. Sequential latent knowledge selec-tion for knowledge-grounded dialogue[J]. arXiv:2002.07510, 2020.
[100] ZHENG C, CAO Y, JIANG D, et al. Difference-aware know-ledge selection for knowledge-grounded conversation gene-ration[J]. arXiv:2009.09378, 2020.
[101] ZHAN H, ZHANG H, CHEN H, et al. Augmenting know-ledge-grounded conversations with sequential knowledge transition[C]//Proceedings of the 2021 Conference of the North American Chapter of the Association for Compu-tational Linguistics: Human Language Technologies. Stroud-sburg: ACL, 2021: 5621-5630.
[102] BENGIO Y, DUCHARME R, VINCENT P, et al. A neural probabilistic language model[J]. The Journal of Machine Learning Research, 2003, 3: 1137-1155.
[103] PAPINENI K, ROUKOS S, WARD T, et al. Bleu: a method for automatic evaluation of machine translation[C]//Pro-ceedings of the 40th Annual Meeting of the Association for Computational Linguistics, Grenoble, Apr 8-9, 2002: 311-318.
[104] LIN C Y. Rouge: a package for automatic evaluation of summaries[C]//Text Summarization Branches Out. Stroud-sburg: ACL, 2004: 74-81.
[105] LI J, GALLEY M, BROCKETT C, et al. A diversity-promoting objective function for neural conversation models[J]. arXiv:1510.03055, 2015.
[106] SEE A, LIU P J, MANNING C D. Get to the point: sum-marization with pointer-generator networks[C]//Proceedings of the 55th Annual Meeting of the Association for Compu-tational Linguistics, Vancouver, Jul 30-Aug 4, 2017: 1073-1083.
[107] ZHOU H, YOUNG T, HUANG M, et al. Commonsense knowledge aware conversation generation with graph attention[C]//Proceedings of the 2018 International Joint Conference on Artificial Intelligence. Stroudsburg: ACL, 2018: 4623-4629.
[108] JUNG J, SON B, LYU S. Attnio: knowledge graph explo-ration with in-and-out attention flow for knowledge-grounded dialogue[C]//Proceedings of the 2020 Confe-rence on Empirical Methods in Natural Language Proces-sing, Hong Kong, China, Nov 16-20, 2020: 3484-3497.
[109] XU J, WANG H, NIU Z Y, et al. Conversational graph grounded policy learning for open-domain conversation generation[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. Stroud-sburg: ACL, 2020: 1835-1845.
[110] MOON S, SHAH P, KUMAR A, et al. OpenDialKG: explainable conversational reasoning with attention-based walks over knowledge graphs[C]//Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. Stroudsburg: ACL, 2019: 845-854.