Elamur Research⌕
☰
home / papers / 2021 / Neural Path Hunter: Reducing Hallucination in Dialogue Systems via Path Grounding

Neural Path Hunter: Reducing Hallucination in Dialogue Systems via Path Grounding

Paper · 2021 · Nouha Dziri, Andrea Madotto, Osmar Zaïane et al.

Summary

Large pre-trained dialogue models produce fluent replies, but they often state facts that are wrong. This paper tackles that problem for dialogue grounded in a Knowledge Graph (KG). The authors first ran a crowd-sourced human study of GPT-2 responses fine-tuned on OpenDialKG. It showed that most errors are extrinsic hallucinations, meaning the model introduces entities that the local KG subgraph does not support.

To fix such errors, they propose NEURAL PATH HUNTER (NPH), a generate-then-refine module that can be attached to any response generator without retraining it. The module works in three steps:

  • A token-level critic flags entity mentions that look suspicious and masks them.
  • A non-autoregressive LM produces contextual representations for the masked slots, which GPT-2 then processes further.
  • The resulting representations query a k-hop subgraph anchored in the conversation history, which returns correct replacement entities.

On OpenDialKG, NPH raises faithfulness substantially according to FeQA and human judgments, with only a small loss in fluency. The work matters because it shows that targeted, KG-based post-editing can make conversational agents more trustworthy.

Key points

  • Human study: annotators judged 1,500 GPT-2 responses, 300 per decoding strategy (greedy, beam, two nucleus settings, top-k), with Krippendorff's alpha of 0.72.
    • Greedy decoding hallucinated least, while top-k reached 40.33%.
    • More diverse decoding coincided with more hallucination.
  • Dataset observation: 54.80% of OpenDialKG responses mention entities that neither the dialogue history nor the 1-hop subgraph supports.
  • Main result: NPH gives a relative FeQA gain of 20.35% over vanilla GPT2 and a 39.98% improvement in human evaluation.
  • Human evaluation details:
    • The critic's precision was 97.5% for GPT2-KB, 95.5% for AdapterBot, and 97.0% for GPT2-KE.
    • For GPT2-KB, hallucinations fell by 42.05%, while fluency dropped by 4.32%.
  • Ablation: representing the local subgraph aligned with GPT2 works better than adding global graph structure to the entity embeddings. Code: github.com/nouhadziri/Neural-Path-Hunter.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓Hong Kong University of Science and TechnologyHong Kong · China58 works · 23,058 citations🎓McGill UniversityMontreal · Canada51 works · 10,476 citations🔬Mila - Quebec Artificial Intelligence InstituteMontreal · Canada31 works · 13,061 citations🔬Canadian Institute for Advanced ResearchToronto · Canada17 works · 16,274 citations🎓University of AlbertaEdmonton · Canada23 works · 2,655 citations

📚 Shelves 4 Computer Science · 9,086 works Artificial Intelligence · 8,889 works Arts and Humanities · 9,296 works Language and Linguistics · 9,206 works

🧭 Topics 1 Dialogue systems · 401 works

Cite

Cite as Nouha Dziri, Andrea Madotto, Osmar Zaïane and Avishek Joey Bose. 2021. Neural Path Hunter: Reducing Hallucination in Dialogue Systems via Path Grounding. Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2021.emnlp-main.168

@inproceedings{dziri2021neural,
  title = {{Neural Path Hunter: Reducing Hallucination in Dialogue Systems via Path Grounding}},
  author = {Nouha Dziri, Andrea Madotto, Osmar Zaïane and Avishek Joey Bose},
  booktitle = {Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing},
  year = {2021},
  doi = {10.18653/v1/2021.emnlp-main.168},
  url = {https://doi.org/10.18653/v1/2021.emnlp-main.168}
}

References

62 references, 20 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Kian Ahrabian, Aarash Feizi, Yasmin Salehi, William L. Hamilton, and Avishek Joey Bose. 2020. Structure aware negative sampling in knowledge graphs. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 6093–6101, Online. Association for Computation… find
  2. Anusha Balakrishnan, Jinfeng Rao, Kartikeya Upasani, Michael White, and Rajen Subba. 2019. Constrained decoding for neural NLG from compositional representations in task-oriented dialogue. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 831– 844, Flo… our card
  3. Ondřej Dušek, Jekaterina Novikova, and Verena Rieser. 2018. Findings of the E2E NLG challenge. In Proceedings of the 11th International Conference on Natural Language Generation, pages 322–328, Tilburg University, The Netherlands. Association for Computational Linguistics. find
  4. Ondřej Dušek, Jekaterina Novikova, and Verena Rieser. 2020. Evaluating the state-of-the-art of end-to-end natural language generation: The e2e nlg challenge. Computer Speech & Language, 59:123–156. find
  5. Nouha Dziri, Ehsan Kamalloo, Kory Mathewson, and Osmar Zaiane. 2019. Augmenting neural response generation with context-aware topical attention. In Proceedings of the First Workshop on NLP for Conversational AI, pages 18–31, Florence, Italy. Association for Computational Linguistics. find
  6. Hannah Bast, Florian Bäurle, Björn Buchhold, and Elmar Haußmann. 2014. Easy access to the freebase dataset. In Proceedings of the 23rd International Conference on World Wide Web, pages 95–98. find
  7. Nouha Dziri, Hannah Rashkin, Tal Linzen, and David Reitter. 2021. Evaluating groundedness in dialogue systems: The begin benchmark. ArXiv preprint, abs/2105.00071. — [arXiv:2105.00071](https://arxiv.org/abs/2105.00071) find
  8. Jonathan Berant, Andrew Chou, Roy Frostig, and Percy Liang. 2013. Semantic parsing on Freebase from question-answer pairs. In Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing, pages 1533–1544, Seattle, Washington, USA. Association for Computational Linguistics.… our card
  9. Falcon, WA. 2019. Pytorch lightning. GitHub. Note: https://github.com/PyTorchLightning/pytorch-lightning, 3. ↗
  10. Kurt Bollacker, Colin Evans, Praveen Paritosh, Tim Sturge, and Jamie Taylor. 2008. Freebase: a collaboratively created graph database for structuring human knowledge. In Proceedings of the 2008 ACM SIGMOD international conference on Management of data, pages 1247–1250. find
  11. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. BERT: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies,… our card
  12. Bhuwan Dhingra, Manaal Faruqui, Ankur Parikh, Ming-Wei Chang, Dipanjan Das, and William Cohen. 2019. Handling divergent reference texts when evaluating table-to-text generation. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 4884–4895, Florence, Ita… our card
  13. Esin Durmus, He He, and Mona Diab. 2020. FEQA: A question answering evaluation framework for faithfulness assessment in abstractive summarization. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 5055– 5070, Online. Association for Computational Lingu… our card
  14. Shi Feng, Eric Wallace, Alvin Grissom II, Mohit Iyyer, Pedro Rodriguez, and Jordan Boyd-Graber. 2018. Pathologies of neural models make interpretations difficult. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 3719–3728, Brussels, Belgium. Associatio… find
  15. Michael Gutmann and Aapo Hyvärinen. 2010. Noise-contrastive estimation: A new estimation principle for unnormalized statistical models. In Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, pages 297–304. find
  16. Yanchao Hao, Yuanzhe Zhang, Kang Liu, Shizhu He, Zhanyi Liu, Hua Wu, and Jun Zhao. 2017. An end-to-end model for question answering over knowledge base with cross-attention combining global knowledge. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume… find
  17. Ari Holtzman, Jan Buys, Li Du, Maxwell Forbes, and Yejin Choi. 2020. The curious case of neural text degeneration. In 8th International Conference on Learning Representations, ICLR 2020, Addis Ababa, Ethiopia, April 26-30, 2020. OpenReview.net. find
  18. Matthew Honnibal and Ines Montani. 2017. spacy 2: Natural language understanding with bloom embeddings, convolutional neural networks and incremental parsing. To appear, 7(1). find
  19. Neil Houlsby, Andrei Giurgiu, Stanislaw Jastrzebski, Bruna Morrone, Quentin de Laroussilhe, Andrea Gesmundo, Mona Attariyan, and Sylvain Gelly. 2019. Parameter-efficient transfer learning for NLP. In Proceedings of the 36th International Conference on Machine Learning, ICML 2019, 9-15 June 2019, Lon… find
  20. Mandar Joshi, Danqi Chen, Yinhan Liu, Daniel S. Weld, Luke Zettlemoyer, and Omer Levy. 2020. SpanBERT: Improving pre-training by representing and predicting spans. Transactions of the Association for Computational Linguistics, 8:64–77. our card
  21. Daniel Jurafsky and James H Martin. 2018. Speech and language processing (draft). Chapter A: Hidden Markov Models (Draft of September 11, 2018). Retrieved March, 19:2019. find
  22. Nitish Shirish Keskar, Bryan McCann, Lav R Varshney, Caiming Xiong, and Richard Socher. 2019. Ctrl: A conditional transformer language model for controllable generation. ArXiv preprint, abs/1909.05858. our card
  23. Diederik P. Kingma and Jimmy Ba. 2015. Adam: A method for stochastic optimization. In 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, May 7-9, 2015, Conference Track Proceedings. find
  24. Thomas N. Kipf and Max Welling. 2017. Semi-supervised classification with graph convolutional networks. In 5th International Conference on Learning Representations, ICLR 2017, Toulon, France, April 24-26, 2017, Conference Track Proceedings. OpenReview.net. our card
  25. Philipp Koehn and Rebecca Knowles. 2017. Six challenges for neural machine translation. In Proceedings of the First Workshop on Neural Machine Translation, pages 28–39, Vancouver. Association for Computational Linguistics. find
  26. Wojciech Kryscinski, Bryan McCann, Caiming Xiong, and Richard Socher. 2020. Evaluating the factual consistency of abstractive text summarization. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 9332–9346, Online. Association for Computational… find
  27. Katherine Lee, Orhan Firat, Ashish Agarwal, Clara Fannjiang, and David Sussillo. 2019. Hallucinations in neural machine translation. In International Conference on Learning Representations. find
  28. Zhibin Liu, Zheng-Yu Niu, Hua Wu, and Haifeng Wang. 2019b. Knowledge aware conversation generation with explainable reasoning over augmented graphs. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Languag… find
  29. Andrea Madotto, Samuel Cahyawijaya, Genta Indra Winata, Yan Xu, Zihan Liu, Zhaojiang Lin, and Pascale Fung. 2020. Learning knowledge bases with parameters for task-oriented dialogue systems. In Findings of the Association for Computational Linguistics: EMNLP 2020, pages 2372–2394, Online. Associatio… find
  30. Joshua Maynez, Shashi Narayan, Bernd Bohnet, and Ryan McDonald. 2020. On faithfulness and factuality in abstractive summarization. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 1906–1919, Online. Association for Computational Linguistics. find
  31. Seungwhan Moon, Pararth Shah, Anuj Kumar, and Rajen Subba. 2019. OpenDialKG: Explainable conversational reasoning with attention-based walks over knowledge graphs. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 845–854, Florence, Italy. Association… our card
  32. Allen Newell, John C Shaw, and Herbert A Simon. 1959. Report on a general problem solving program. In IFIP congress, volume 256, page 64. Pittsburgh, PA. find
  33. Allen Newell and Herbert Simon. 1956. The logic theory machine–a complex information processing system. IRE Transactions on information theory, 2(3):61–79. find
  34. Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. 2002. Bleu: a method for automatic evaluation of machine translation. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, pages 311–318, Philadelphia, Pennsylvania, USA. Association for Computational… our card
  35. Zhaojiang Lin, Andrea Madotto, Yejin Bang, and Pascale Fung. 2020. The adapter-bot: All-in-one controllable conversational model. ArXiv preprint, abs/2008.12579. — [arXiv:2008.12579](https://arxiv.org/abs/2008.12579) find
  36. Ankur Parikh, Xuezhi Wang, Sebastian Gehrmann, Manaal Faruqui, Bhuwan Dhingra, Diyi Yang, and Dipanjan Das. 2020. ToTTo: A controlled table-to-text generation dataset. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 1173–1186, Online. Associat… our card
  37. Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. 2019a. Roberta: A robustly optimized bert pretraining approach. ArXiv preprint, abs/1907.11692. find
  38. Fabio Petroni, Tim Rocktäschel, Sebastian Riedel, Patrick Lewis, Anton Bakhtin, Yuxiang Wu, and Alexander Miller. 2019. Language models as knowledge bases? In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural… our card
  39. Ran Tian, Shashi Narayan, Thibault Sellam, and Ankur P Parikh. 2019. Sticking to the facts: Confident decoding for faithful data-to-text generation. ArXiv preprint, abs/1910.08684. — [arXiv:1910.08684](https://arxiv.org/abs/1910.08684) find
  40. Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, and Ilya Sutskever. 2019. Language models are unsupervised multitask learners. OpenAI Blog. find
  41. Yi-Lin Tuan, Yun-Nung Chen, and Hung-yi Lee. 2019. DyKgChat: Benchmarking dialogue generation grounding on dynamic knowledge graphs. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EM… find
  42. Hannah Rashkin, David Reitter, Gaurav Singh Tomar, and Dipanjan Das. 2021. Increasing faithfulness in knowledge-grounded dialogue with controllable features. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natu… our card
  43. Adam Roberts, Colin Raffel, and Noam Shazeer. 2020. How much knowledge can you pack into the parameters of a language model? In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 5418–5426, Online. Association for Computational Linguistics. find
  44. Anna Rohrbach, Lisa Anne Hendricks, Kaylee Burns, Trevor Darrell, and Kate Saenko. 2018. Object hallucination in image captioning. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 4035–4045, Brussels, Belgium. Association for Computational Linguistics. our card
  45. Iulian Vlad Serban, Alessandro Sordoni, Yoshua Bengio, Aaron C. Courville, and Joelle Pineau. 2016. Building end-to-end dialogue systems using generative hierarchical neural network models. In Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence, February 12-17, 2016, Phoenix, Ari… our card
  46. Amit Singhal. 2012. Introducing the knowledge graph: things, not strings. Official google blog, 5. find
  47. Alessandro Sordoni, Michel Galley, Michael Auli, Chris Brockett, Yangfeng Ji, Margaret Mitchell, Jian-Yun Nie, Jianfeng Gao, and Bill Dolan. 2015. A neural network approach to context-sensitive generation of conversational responses. In Proceedings of the 2015 Conference of the North American Chapte… our card
  48. Nitish Srivastava, Geoffrey Hinton, Alex Krizhevsky, Ilya Sutskever, and Ruslan Salakhutdinov. 2014. Dropout: a simple way to prevent neural networks from overfitting. The journal of machine learning research, 15(1):1929–1958. find
  49. Wilson L Taylor. 1953. “cloze procedure”: A new tool for measuring readability. Journalism quarterly, 30(4):415–433. find
  50. Shikhar Vashishth, Soumya Sanyal, Vikram Nitin, and Partha P. Talukdar. 2020. Composition-based multi-relational graph convolutional networks. In 8th International Conference on Learning Representations, ICLR 2020, Addis Ababa, Ethiopia, April 26-30, 2020. OpenReview.net. find
  51. Alex Wang, Kyunghyun Cho, and Mike Lewis. 2020. Asking and answering questions to evaluate the factual consistency of summaries. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 5008–5020, Online. Association for Computational Linguistics. find
  52. Zhen Wang, Jianwen Zhang, Jianlin Feng, and Zheng Chen. 2014. Knowledge graph embedding by translating on hyperplanes. In Proceedings of the Twenty-Eighth AAAI Conference on Artificial Intelligence, July 27 -31, 2014, Québec City, Québec, Canada, pages 1112–1119. AAAI Press. find
  53. Ronald J Williams and David Zipser. 1989. Experimental analysis of the real-time recurrent learning algorithm. Connection science, 1(1):87–111. find
  54. Sam Wiseman, Stuart Shieber, and Alexander Rush. 2017. Challenges in data-to-document generation. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 2253–2263, Copenhagen, Denmark. Association for Computational Linguistics. our card
  55. Thomas Wolf, Lysandre Debut, Victor Sanh, Julien Chaumond, Clement Delangue, Anthony Moi, Pier-ric Cistac, Tim Rault, Remi Louf, Morgan Funtow-icz, Joe Davison, Sam Shleifer, Patrick von Platen, Clara Ma, Yacine Jernite, Julien Plu, Canwen Xu, Teven Le Scao, Sylvain Gugger, Mariama Drame, Quentin Lh… our card
  56. Hu Xu, Seungwhan Moon, Honglei Liu, Bing Liu, Pararth Shah, Bing Liu, and Philip Yu. 2020. User memory reasoning for conversational recommendation. In Proceedings of the 28th International Conference on Computational Linguistics, pages 5288– 5308, Barcelona, Spain (Online). International Committee o… find
  57. Xuchen Yao and Benjamin Van Durme. 2014. Information extraction over structured data: Question answering with Freebase. In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 956–966, Baltimore, Maryland. Association for Computationa… find
  58. Tom Young, Erik Cambria, Iti Chaturvedi, Hao Zhou, Subham Biswas, and Minlie Huang. 2018. Augmenting end-to-end dialogue systems with commonsense knowledge. In Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence, (AAAI-18), the 30th innovative Applications of Artificial Intel… find
  59. Mo Yu and Mark Dredze. 2014. Improving lexical embeddings with semantic knowledge. In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 545–550, Baltimore, Maryland. Association for Computational Linguistics. our card
  60. Yizhe Zhang, Siqi Sun, Michel Galley, Yen-Chun Chen, Chris Brockett, Xiang Gao, Jianfeng Gao, Jingjing Liu, and Bill Dolan. 2020. DIALOGPT : Large-scale generative pre-training for conversational response generation. In Proceedings of the 58th Annual Meeting of the Association for Computational Ling… our card
  61. Hao Zhou, Tom Young, Minlie Huang, Haizhou Zhao, Jingfang Xu, and Xiaoyan Zhu. 2018. Commonsense knowledge aware conversation generation with graph attention. In Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence, IJCAI 2018, July 13-19, 2018, Stockholm, Swed… find
  62. Wenya Zhu, Kaixiang Mo, Yu Zhang, Zhangbin Zhu, Xuezheng Peng, and Qiang Yang. 2017. Flexible end-to-end dialogue system for knowledge grounded conversation. arXiv, pages arXiv–1709. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/dziri-2021-neural-path-hunter-reducing-hallucination/