Skip to main content
SpringerLink
Log in

GraphFlow+: Exploiting Conversation Flow in Conversational Machine Comprehension with Graph Neural Networks

  • Research Article
  • Published:
Machine Intelligence Research Aims and scope Submit manuscript

Abstract

The conversation machine comprehension (MC) task aims to answer questions in the multi-turn conversation for a single passage. However, recent approaches don’t exploit information from historical conversations effectively, which results in some references and ellipsis in the current question cannot be recognized. In addition, these methods do not consider the rich semantic relationships between words when reasoning about the passage text. In this paper, we propose a novel model GraphFlow+, which constructs a context graph for each conversation turn and uses a unique recurrent graph neural network (GNN) to model the temporal dependencies between the context graphs of each turn. Specifically, we exploit three different ways to construct text graphs, including the dynamic graph, static graph, and hybrid graph that combines the two. Our experiments on CoQA, QuAC and DoQA show that the GraphFlow+ model can outperform the state-of-the-art approaches.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  1. L. F. Wu, Y. Chen, K. Shen, X. J. Guo, H. N. Gao, S. C. Li, J. Pei, B. Long. Graph neural networks for natural language processing: A survey, [Online], Available: https://arxiv.org/abs/2106.06090, 2021.

  2. P. Rajpurkar, J. Zhang, K. Lopyrev, P. Liang. SQuAD: 100 000+ questions for machine comprehension of text. In Proceedings of Conference on Empirical Methods in Natural Language Processing, Austin, USA, pp.2383-2392, 2016. DOI: https://doi.org/10.18653/v1/D16-1264.

  3. E. Choi, H. He, M. Iyyer, M. Yatskar, W. T. Yih, Y. Choi, P. Liang, L. Zettlemoyer. QuAC: Question answering in context In Proceedings of Conference on Empirical Methods in Natural Language Processing, Brussels, Belgium, pp. 2174–2184, 2018. DOI: https://doi.org/10.18653/v1/D18-1241.

  4. S. Reddy, D. Q. Chen, C. D. Manning. CoQA: A conversational question answering challenge. Transactions of the Association for Computational Linguistics, vol. 7, pp. 249–266, 2019. DOI: https://doi.org/10.1162/tacl_a_00266.

    Article  Google Scholar 

  5. C. G. Zhu, M. Zeng, X. D. Huang. SDNet: Contextualized attention-based deep network for conversational question answering, [Online], Available: https://arxiv.org/abs/1812.03593, 2018.

  6. M. Yatskar. A qualitative comparison of CoQA, squad 2.0 and QuAC. In Proceedings of Conference of the North American Chapter of the Association for Computational Linguistics:: Human Language Technologies, Minneapolis, USA, pp. 2318–2323, 2019. DOI: https://doi.org/10.18653/v1/N19-1241.

  7. H. Y. Huang, E. Choi, W. T. Yih. FlowQA: Grasping flow in history for conversational machine comprehension. In Proceedings of the 7th International Conference on Learning Representations, New Orleans, USA, 2019.

  8. N. De Cao, W. Aziz, I. Titov. Question answering by reasoning across documents with graph convolutional networks. In Proceedings of Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Minneapolis, USA, pp. 2306–2317, 2019. DOI: https://doi.org/10.18653/v1/N19-1240.

  9. L. F. Song, Z. G. Wang, M. Yu, Y. Zhang, R. Florian, D. Gildea. Exploring graph-structured passage representation for multi-hop reading comprehension with graph neural networks, [Online], Available: https://arxiv.org/abs/1809.02040, 2018.

  10. Y. Chen, L. F. Wu, M. J. Zaki. GraphFlow: Exploiting conversation flow with graph neural networks for conversational machine comprehension. In Proceedings of the 29th International Joint Conference on Artificial Intelligence, Yokohama, Japan, pp. 1230–1236, 2020.

  11. L. F. Wu, P. Cui, J. Pei, L. Zhao. Graph Neural Networks: Foundations, Frontiers, and Applications, Singapore, Singapore: Springer, 2022. DOI: https://doi.org/10.1007/978-981-16-6054-2.

    Book  Google Scholar 

  12. J. Pennington, R. Socher, C. Manning. GloVe: Global vectors for word representation. In Proceedings of Conference on Empirical Methods in Natural Language Processing, Doha, Qatar, pp. 1532–1543, 2014. DOI: https://doi.org/10.3115/v1/D14-1162.

  13. J. Devlin, M. W. Chang, K. Lee, K. Toutanova. BERT: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Minneapolis, USA, pp.4171-4186, 2019. DOI: https://doi.org/10.18653/vl/N19-1423.

  14. K. Lee, S. Salant, T. Kwiatkowski, A. Parikh, D. Das, J. Berant. Learning recurrent span representations for extractive question answering, [Online], Available: https://arxiv.org/abs/1611.01436, 2016.

  15. K. Xu, L. F. Wu, Z. G. Wang, M. Yu, L. W. Chen, V. Sheinin Exploiting rich syntactic information for semantic parsing with graph-to-sequence model In Proceedings of Conference on Empirical Methods in Natural Language Processing, Brussels, Belgium, pp. 918–924, 2018 DOI: https://doi.org/10.18653/v1/D18-1110.

  16. Y. Chen, L. F. Wu, M. Zaki. Iterative deep graph learning for graph neural networks: Better and robust node embeddings. In Proceedings of the 34th International Conference on Neural Information Processing Systems, Vancouver, Canada, pp. 19314–19326, 2020.

  17. S. Hochreiter, J. Schmidhuber. Long short-term memory. Neural Computation, vol. 9, no. 8, pp. 1735–1780, 1997. DOI: https://doi.org/10.1162/neco.1997.9.8.1735.

    Article  CAS  PubMed  Google Scholar 

  18. Y. J. Li, D. Tarlow, M. Brockschmidt, R. S. Zemel. Gated graph sequence neural networks, [Online], Available: https://arxiv.org/abs/1511.05493, 2015.

  19. K. Cho, B. van Merriënboer, C. Gulcehre, D. Bahdanau, F. Bougares, H. Schwenk, Y. Bengio. Learning phrase representations using RNN encoder-decoder for statistical machine translation. In Proceedings of Conference on Empirical Methods in Natural Language Processing, Doha, Qatar, pp. 1724–1734, 2014. DOI: https://doi.org/10.3115/v1/D14-1179.

  20. W. Wang, M. Yan, C. Wu. Multi-granularity hierarchical attention fusion networks for reading comprehension and question answering. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, Melbourne, Australia, pp. 1705–1714, 2018. DOI: https://doi.org/10.18653/v1/P18-1158.

  21. J. A. Campos, A. Otegi, A. Soroa, J. Deriu, M. Cieliebak, E. Agirre. Conversational QA for FAQs. In Proceedings of the 33rd Conference on Neural Information Processing Systems, Vancouver, Canada, 2019.

  22. A. See, P. J. Liu, C. D. Manning. Get to the point: Summarization with pointer-generator networks. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, Vancouver, Canada, pp. 1073–1083, 2017. DOI: https://doi.org/10.18653/v1/P17-1099.

  23. D. Q. Chen, A. Fisch, J. Weston, A. Bordes. Reading Wikipedia to answer open-domain questions. In Proceedingsof the 55th Annual Meeting of the Association for Computational Linguistics, Vancouver, Canada, pp. 1870–1879, 2017. DOI: https://doi.org/10.18653/v1/P17-1171.

  24. C. Manning, M. Surdeanu, J. Bauer, J. Finkel, S. Bethard, D. McClosky. The Stanford CoreNLP natural language processing toolkit. In Proceedings of 52nd Annual Meeting of the Association for Computational Linguistics: System Demonstrations, Baltimore, USA, pp.55-60, 2014. DOI: https://doi.org/10.3115/v1/P14-5010.

  25. D. P. Kingma, J. Ba. Adam: A method for stochastic optimization, [Online], Available: https://arxiv.org/abs/1412.6980, 2014.

  26. M. Seo, A. Kembhavi, A. Farhadi, H. Hajishirzi. Bidirectional attention flow for machine comprehension. In Proceedings of the 5th International Conference on Learning Representations, Toulon, France, 2017.

  27. C. M. Xiong, V. Zhong, R. Socher. Dynamic coattention networks for question answering. In Proceedings of the 5th International Conference on Learning Representations, Toulon, France, 2017.

  28. Y. Ju, F. B. Zhao, S. J. Chen, B. W. Zheng, X. F. Yang, Y. F. Liu. Technical report on conversational question answering, [Online], Available: https://arxrv.org/abs/1909.10772, 2019.

  29. C. Qu, L. Yang, M. H. Qiu, Y. F. Zhang, C. Chen, W. B. Croft, M. Iyyer. Attentive history selection for conversational question answering. In Proceedings of the 28th ACM International Conference on Information and Knowledge Managemen, Beijing, China, pp. 1391–1400, 2019. DOI: https://doi.org/10.1145/3357384.3357905.

  30. Y. T. Yeh, Y. N. Chen. FlowDelta: Modeling flow information gain in reasoning for conversational machine comprehension. In Proceedings of the 2nd Workshop on Machine Reading for Question Answering, Hong Kong, China, pp. 86–90, 2019. DOI: https://doi.org/10.18653/v1/D19-5812.

  31. K. Xu, L. F. Wu, Z. G. Wang, V. Sheinin. Graph2Seq: Graph to sequence learning with attention-based neural networks, [Online], Available: https://arxiv.org/abs/1804.00823, 2018.

  32. J. Gilmer, S. S. Schoenholz, P. F. Riley, O. Vinyals, G. E. Dahl. Neural message passing for Quantum chemistry. In Proceedings of the 34th International Conference on Machine Learning, Sydney, Australia, pp. 1263–1272, 2017.

  33. W. L. Hamilton, R. Ying, J. Leskovec. Inductive representation learning on large graphs. In Proceedings of the 31st International Conference on Neural Information Processing Systems, Long Beach, USA, pp. 1025–1035, 2017.

  34. T. N. Kipf, M. Welling. Semi-supervised classification with graph convolutional networks, [Online], Available: https://arxiv.org/abs/1609.02907, 2016.

  35. H. T. Sun, B. Dhigula, M. Zaheer, K. Mazaitis, R. Salakhutdinov, W. Cohen. Open domain question answering using early fusion of knowledge bases and text. In Proceedings of Conference on Empirical Methods in Natural Language Processing, Brussels, Belgium, pp. 4231–4242, 2018. DOI: https://doi.org/10.18653/v1/D18-1455.

  36. Y. Chen, L. F. Wu, M. J. Zaki. Toward subgraph guided knowledge graph question generation with graph neural networks, [Online], Available: https://arxiv.org/abs/2004.06015, 2020.

  37. Q. Ran, Y. K. Lin, P. Li, J. Zhou, Z. Y. Liu. NumNet: Machine reading comprehension with numerical reasoning. In Proceedings of Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, Hong Kong, China, pp. 2474–2484, 2019. DOI: https://doi.org/10.18653/v1/D19-1251.

  38. C. Zheng, P. Kordjamshidi. SRLGRN: Semantic role labeling graph reasoning network. In Proceedings of Conference on Empirical Methods in Natural Language Processing, pp. 8881–8891, 2020. DOI: https://doi.org/10.18653/v1/2020.emnlp-main.714.

  39. P. F. Liu, S. C. Chang, X. J. Huang, J. Tang, J. C. K. Cheung. Contextualized non-local neural networks for sequence learning. In Proceedings of the 31st AAAI Conference on Artificial Intelligence and the 31st Innovative Applications of Artificial Intelligence Conference and the 9th AAAI Symposium on Educational Advances in Artificial Intelligence, Honolulu, USA, pp. 830, 2018. DOI: https://doi.org/10.1609/aaai.v33i01.33016762.

  40. Y. Chen, L. F. Wu, M. J. Zaki. Reinforcement learning based graph-to-sequence model for natural question generation. In Proceedings of the 8th International Conference on Learning Representations, Addis Ababa, Ethiopia, 2020.

Download references

Acknowledegments

This article is a substantial extension of our earlier work presented at the 29th International Joint Conference on Artificial Intelligence and the 17th Pacific Rim International Conference on Artificial Intelligence (IJCAI-PRICAI-20) in 2020.

Author information

Author notes

  1. These authors contribute equally to this work

Authors and Affiliations

  1. School of Computer Science, Central China Normal University, Wuhan, 430079, China

    Jing Hu & Po Hu

  2. Pinterest, San Francisco, 94016, USA

    Lingfei Wu

  3. Meta, Mountain View, 94039, USA

    Yu Chen

  4. Rensselaer Polytechnic Institute, Troy, 12180, USA

    Mohammed J. Zaki

Authors
  1. Jing Hu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Lingfei Wu
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Yu Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Po Hu
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Mohammed J. Zaki
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Lingfei Wu.

Ethics declarations

The authors declared that they have no conflicts of interest to this work.

Additional information

Colored figures are available in the online version at https://link.springer.com/journal/11633

Jing Hu received the B. Sc. degree in software engineer from Central China Normal University, China in 2023. Currently, she is a master student in computer technology at Tsinghua Shenzhen International Graduate School, Tsinghua University, China.

Her research interests include natural language processing, pretrained language models and code summarization.

Lingfei Wu received the Ph.D. degree in computer science from College of William and Mary, USA in 2016. Currently, he is an engineering manager in the Content and Knowledge Graph Group at Pinterest, USA. He has published one book (in GNNs) and more than 100 top-ranked AI/ML/NLP conferences and journal papers, including but not limited to NeurIPS, ICML, ICLR, KDD, ACL, EMNLP, NAACL, IJCAI, and AAAI. He is also a co-inventor of more than 40 filed US patents. Because of the commercial value of his patents, he received several invention achievement awards and was appointed as IBM Master Inventors, class of 2020. He was the recipient of the Best Paper Award and Best Student Paper Award of several conferences such as IEEE ICC19, AAAI workshop on DLGMA20 and KDD workshop on DLG19.

His research interests include the intersection of machine learning (deep learning), representation learning, and natural language processing, with a particular emphasis on the fast-growing subjects of graph neural networks and its extensions on new application domains.

Yu Chen received the Ph.D. degree in computer science from Rensselaer Polytechnic Institute, USA in 2020. Currently, he is a senior research scientist at Meta AI, USA.

His research interests include the intersection of machine learning (deep learning) and natural language processing, with a particular emphasis on the fast-growing field of graph neural networks and their applications in various domains.

Po Hu received the Ph.D. degree in computer software and theory from Wuhan University, China in 2013. He was a visiting scholar at Hong Kong Baptist University, China. Currently, he is an associate professor of computer science at Central China Normal University, China. He has published over 30 papers in prestigious journals and prominent conferences (e.g., IJCAI, ACL, COLING). He has served as the program committee/journal reviewer in major AI and NLP conferences (e.g., AAAI, ACL, EMNLP, TKDE).

His research interests include natural language processing, knowledge engineering and computational social science.

Mohammed J. Zaki received the Ph.D. degree in computer science from University of Rochester, USA in 1998. Currently, he is a professor and department head of computer science at Rensselaer Polytechnic Institute, USA.

His research interests include novel data mining and machine learning techniques, particularly for learning from graph structured and textual data, with applications in bioinformatics, personal health and financial analytics.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Hu, J., Wu, L., Chen, Y. et al. GraphFlow+: Exploiting Conversation Flow in Conversational Machine Comprehension with Graph Neural Networks. Mach. Intell. Res. 21, 272–282 (2024). https://doi.org/10.1007/s11633-023-1421-0

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11633-023-1421-0

Keywords

Search

Navigation