Elamur Research
☰
home / papers / 2022 / TRUE: Re-evaluating Factual Consistency Evaluation

TRUE: Re-evaluating Factual Consistency Evaluation

Written for the Research library (claude-opus-5-5) from the paper text. Original — aclanthology.org/2022.dialdoc-1.19.pdf

Summary

Text generation systems that rely on a grounding source often produce content that contradicts or goes beyond that source. Automatic detectors of such factual inconsistency exist, but each is usually built and tested on one task or dataset. They are also typically judged by system-level correlation with human ratings, which says little about how they behave on single outputs. TRUE addresses this by gathering 11 existing human-annotated English datasets into one common format. The datasets cover summarization, knowledge-grounded dialogue, fact verification and paraphrasing. Each example pairs a target text with its grounding text and carries a binary label: consistent or not. On this collection, the authors compare 12 metrics at the example level using ROC AUC for detecting inconsistent outputs. Methods based on large-scale NLI and on question generation plus answering (QG-QA) perform best across tasks. Combining the two improves results further, which suggests there is still headroom. The study gives developers a shared benchmark, an easy-to-interpret protocol and strong baselines for future work.

Key points

  • Datasets:
    • Summarization: FRANK, SummEval, MNBM, QAGS-CNNDM and QAGS-XSum.
    • Dialogue: BEGIN, Q², DialFact.
    • Fact verification: FEVER, VitaminC.
    • Paraphrasing: PAWS.
    • Sizes range from 235 to 63,054 examples, and graded or sentence-level labels were converted to a binary scheme.
  • Strict definition of consistency: a text counts as consistent only if it agrees with its grounding text, regardless of real-world truth. Opinions and chit-chat are left out of scope.
  • Evaluation protocol: ROC AUC replaces correlations, so precision and recall of inconsistency detection can be read directly. NLI-based and QG-QA-based approaches significantly outperform the other metrics.
  • Error analysis:
    • The best metrics missed subtle inconsistencies in 21 of 180 analysed cases.
    • Personal statements in dialogue caused false alarms in 10 of 62 responses.
    • Long inputs are another weak spot.
    • Of 100 sampled examples, 10 had annotation errors.
  • Recommendations:
    • Use NLI and QG-QA as default metrics and as baselines.
    • Report ROC AUC.
    • Data curators should provide a mapping from their labels to binary labels.

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

Fields: Natural language processing

Cited by

215 papers of our library cite it (of the 140673 whose bibliographies we read); 11 of them have a page here, 204 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2022.naacl-main.287

@inproceedings{honovich2022true,
  title = {{TRUE: Re-evaluating Factual Consistency Evaluation}},
  author = {Honovich, Or and Aharoni, Roee and Herzig, Jonathan and Taitelbaum, Hagai and Kukliansy, Doron and Cohen, Vered and Scialom, Thomas and Szpektor, Idan and Hassidim, Avinatan and Matias, Yossi},
  booktitle = {Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies},
  year = {2022},
  doi = {10.18653/v1/2022.naacl-main.287},
  url = {https://doi.org/10.18653/v1/2022.naacl-main.287}
}

References

68 references, 22 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Samuel R. Bowman, Gabor Angeli, Christopher Potts, and Christopher D. Manning. 2015. A large annotated corpus for learning natural language inference. In Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, pages 632–642, Lisbon, Portugal. Association for Computati… find
  2. Gillian Brown and George Yule. 1983. Discourse Analysis. Cambridge University Press. find
  3. Ido Dagan, Oren Glickman, and Bernardo Magnini. 2006. The pascal recognising textual entailment challenge. In Machine Learning Challenges. Evaluating Predictive Uncertainty, Visual Object Classification, and Recognising Tectual Entailment, pages 177–190, Berlin, Heidelberg. Springer Berlin Heidelber… find
  4. Emily Denton, Mark Díaz, Ian Kivlichan, Vinodkumar Prabhakaran, and Rachel Rosen. 2021. Whose ground truth? accounting for individual and collective identities underlying dataset annotation. arXiv preprint arXiv:2112.04554. — [arXiv:2112.04554](https://arxiv.org/abs/2112.04554) find
  5. Daniel Deutsch, Tania Bedrax-Weiss, and Dan Roth. 2021. Towards question-answering as an automatic metric for evaluating the content quality of a summary. Transactions of the Association for Computational Linguistics, 9:774–789. find
  6. 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,… find
  7. Emily Dinan, Stephen Roller, Kurt Shuster, Angela Fan, Michael Auli, and Jason Weston. 2019. Wizard of Wikipedia: Knowledge-powered conversational agents. In Proceedings of the International Conference on Learning Representations (ICLR). find
  8. 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… find our card
  9. Nouha Dziri, Hannah Rashkin, Tal Linzen, and David Reitter. 2021. Evaluating groundedness in dialogue systems: The begin benchmark. find
  10. Bradley Efron. 1982. The jackknife, the bootstrap and other resampling plans. SIAM. find
  11. Alexander R Fabbri, Wojciech Kryściński, Bryan McCann, Caiming Xiong, Richard Socher, and Dragomir Radev. 2020. Summeval: Re-evaluating summarization evaluation. arXiv preprint arXiv:2007.12626. find
  12. Alexander R Fabbri, Wojciech Kryściński, Bryan McCann, Caiming Xiong, Richard Socher, and Dragomir Radev. 2021. Summeval: Re-evaluating summarization evaluation. Transactions of the Association for Computational Linguistics, 9:391–409. find
  13. Tobias Falke, Leonardo F. R. Ribeiro, Prasetya Ajie Utama, Ido Dagan, and Iryna Gurevych. 2019. Ranking generated summaries by correctness: An interesting but challenging application for natural language inference. In Proceedings of the 57th Annual Meeting of the Association for Computational Lingui… find our card
  14. C. Fillmore. 1976. Frame semantics and the nature of language *. Annals of the New York Academy of Sciences, 280. find
  15. Markus Freitag, George Foster, David Grangier, Viresh Ratnakar, Qijun Tan, and Wolfgang Macherey. 2021. Experts, errors, and context: A large-scale study of human evaluation for machine translation. arXiv preprint arXiv:2104.14478. find our card
  16. Saadia Gabriel, Asli Celikyilmaz, Rahul Jha, Yejin Choi, and Jianfeng Gao. 2021. GO FIGURE: A meta evaluation of factuality in summarization. In Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, pages 478–487, Online. Association for Computational Linguistics. find
  17. Sebastian Gehrmann, Tosin Adewumi, Karmanya Aggarwal, Pawan Sasanka Ammanamanchi, Anuoluwapo Aremu, Antoine Bosselut, Khyathi Raghavi Chandu, Miruna-Adriana Clinciu, Dipanjan Das, Kaustubh Dhole, Wanyu Du, Esin Durmus, Ondřej Dušek, Chris Chinenye Emezue, Varun Gangal, Cristina Garbacea, Tatsunori… find
  18. Zhenzhong Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel, Piyush Sharma, and Radu Soricut. 2019. Albert: A lite bert for self-supervised learning of language representations. arXiv preprint arXiv:1909.11942. find
  19. In Proceedings of the 1st Workshop on Natural Language Generation, Evaluation, and Metrics (GEM 2021), pages 96–120, Online. Association for Computational Linguistics. find
  20. Prakhar Gupta, Chien-Sheng Wu, Wenhao Liu, and Caiming Xiong. 2021. Dialfact: A benchmark for fact-checking in dialogue. arXiv preprint arXiv:2110.08222. find
  21. Katherine Lee, Orhan Firat, Ashish Agarwal, Clara Fannjiang, and David Sussillo. 2018. Hallucinations in neural machine translation. find
  22. NeurIPS 2018 Workshop on Interpretability and Robustness for Audio, Speech, and Language. find
  23. Pengcheng He, Xiaodong Liu, Jianfeng Gao, and Weizhu Chen. 2021. Deberta: Decoding-enhanced bert with disentangled attention. In International Conference on Learning Representations. find our card
  24. Martin Heidegger. 2001. On the essence of truth. The Nature of Truth: Classic and Contemporary Perspectives, 1:295–316. find
  25. Karl Moritz Hermann, Tomas Kocisky, Edward Grefenstette, Lasse Espeholt, Will Kay, Mustafa Suleyman, and Phil Blunsom. 2015. Teaching machines to read and comprehend. In Advances in Neural Information Processing Systems, volume 28. Curran Associates, Inc. find our card
  26. Or Honovich, Leshem Choshen, Roee Aharoni, Ella Neeman, Idan Szpektor, and Omri Abend. 2021. q 2 : Evaluating factual consistency in knowledge-grounded dialogues via question generation and question answering. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing,… find
  27. J. Edward Hu, Abhinav Singh, Nils Holzenberger, Matt Post, and Benjamin Van Durme. 2019. Large-scale, diverse, paraphrastic bitexts via sampling and clustering. In Proceedings of the 23rd Conference on Computational Natural Language Learning (CoNLL), pages 44–54, Hong Kong, China. Association for Co… find
  28. Tomáš Kočiský, Jonathan Schwarz, Phil Blunsom, Chris Dyer, Karl Moritz Hermann, Gábor Melis, and Edward Grefenstette. 2018. The NarrativeQA reading comprehension challenge. Transactions of the Association for Computational Linguistics, 6:317– 328. find our card
  29. 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
  30. Philippe Laban, Tobias Schnabel, Paul N Bennett, and Marti A Hearst. 2021. Summac: Re-visiting nli-based models for inconsistency detection in summarization. arXiv preprint arXiv:2111.09525. find
  31. Mike Lewis, Yinhan Liu, Naman Goyal, Marjan Ghazvininejad, Abdelrahman Mohamed, Omer Levy, Veselin Stoyanov, and Luke Zettlemoyer. 2020. BART: Denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension. In Proceedings of the 58th Annual Meeting of the… find our card
  32. Chin-Yew Lin. 2004. ROUGE: A package for automatic evaluation of summaries. In Text Summarization Branches Out, pages 74–81, Barcelona, Spain. Association for Computational Linguistics. find
  33. Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. 2019. Roberta: A robustly optimized bert pretraining approach. arXiv preprint arXiv:1907.11692. find
  34. 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
  35. Nikita Nangia, Adina Williams, Angeliki Lazaridou, and Samuel Bowman. 2017. find
  36. The RepEval 2017 shared task: Multi-genre natural language inference with sentence representations. In Proceedings of the 2nd Workshop on Evaluating Vector Space Representations for NLP, pages 1–10, Copenhagen, Denmark. Association for Computational Linguistics. find
  37. Shashi Narayan, Shay B. Cohen, and Mirella Lapata. 2018. Don’t give me the details, just the summary! topic-aware convolutional neural networks for extreme summarization. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 1797–1807, Brussels, Belgium. As… find
  38. Yixin Nie, Haonan Chen, and Mohit Bansal. 2019. Combining fact extraction and verification with neural semantic matching networks. In Association for the Advancement of Artificial Intelligence (AAAI). find
  39. Yixin Nie, Adina Williams, Emily Dinan, Mohit Bansal, Jason Weston, and Douwe Kiela. 2020. Adversarial NLI: A new benchmark for natural language understanding. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 4885–4901, Online. Association for Computa… find
  40. Yixin Nie, Mary Williamson, Mohit Bansal, Douwe Kiela, and Jason Weston. 2021. I like fish, especially dolphins: Addressing contradictions in dialogue modeling. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on N… find
  41. Artidoro Pagnoni, Vidhisha Balachandran, and Yulia Tsvetkov. 2021. Understanding factuality in abstractive summarization with FRANK: A benchmark for factuality metrics. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Languag… find our card
  42. Martha Palmer, Daniel Gildea, and Paul Kingsbury. 2005. The Proposition Bank: An Annotated Corpus of Semantic Roles. Computational Linguistics, 31(1):71–106. find our card
  43. Richard Yuanzhe Pang, Alicia Parrish, Nitish Joshi, Nikita Nangia, Jason Phang, Angelica Chen, Vishakh Padmakumar, Johnny Ma, Jana Thompson, He He, et al. 2021. Quality: Question answering with long input texts, yes! arXiv preprint arXiv:2112.08608. find our card
  44. 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… find our card
  45. 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… find our card
  46. Amy Pu, Hyung Won Chung, Ankur Parikh, Sebastian Gehrmann, and Thibault Sellam. 2021. Learning compact metrics for MT. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 751–762, Online and Punta Cana, Dominican Republic. Association for Computational Li… find our card
  47. Libo Qin, Tianbao Xie, Shijue Huang, Qiguang Chen, Xiao Xu, and Wanxiang Che. 2021. Don’t be contradicted with anything! CI-ToD: Towards benchmarking consistency for task-oriented dialogue system. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 2357–2… find
  48. Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, and Peter J. Liu. 2020. Exploring the limits of transfer learning with a unified text-to-text transformer. Journal of Machine Learning Research, 21(140):1–67. find
  49. Hannah Rashkin, Vitaly Nikolaev, Matthew Lamm, Michael Collins, Dipanjan Das, Slav Petrov, Gaurav Singh Tomar, Iulia Turc, and David Reitter. 2021a. Measuring attribution in natural language generation models. arXiv preprint arXiv:2112.12870. find our card
  50. Hannah Rashkin, David Reitter, Gaurav Singh Tomar, and Dipanjan Das. 2021b. 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 Nat… find
  51. Ehud Reiter and Craig Thomson. 2020. Shared task on evaluating accuracy. In Proceedings of the 13th International Conference on Natural Language Generation, pages 227–231, Dublin, Ireland. Association for Computational Linguistics. find
  52. 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. find our card
  53. Tal Schuster, Adam Fisch, and Regina Barzilay. 2021. Get your vitamin C! robust fact verification with contrastive evidence. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 624–643, Online. Assoc… find our card
  54. Thomas Scialom, Paul-Alexis Dray, Sylvain Lamprier, Benjamin Piwowarski, Jacopo Staiano, Alex Wang, and Patrick Gallinari. 2021. QuestEval: Summarization asks for fact-based evaluation. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 6594–6604, Online… find our card
  55. Thomas Scialom and Felix Hill. 2021. Beametrics: A benchmark for language generation evaluation evaluation. arXiv preprint arXiv:2110.09147. — [arXiv:2110.09147](https://arxiv.org/abs/2110.09147) find
  56. Thibault Sellam, Dipanjan Das, and Ankur Parikh. 2020a. BLEURT: Learning robust metrics for text generation. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 7881–7892, Online. Association for Computational Linguistics. find
  57. Thibault Sellam, Amy Pu, Hyung Won Chung, Sebastian Gehrmann, Qijun Tan, Markus Freitag, Dipanjan Das, and Ankur Parikh. 2020b. Learning to evaluate translation beyond English: BLEURT submissions to the WMT metrics 2020 shared task. In Proceedings of the Fifth Conference on Machine Translation, page… find
  58. Yuan Zhang, Jason Baldridge, and Luheng He. 2019. PAWS: Paraphrase adversaries from word scrambling. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), pages 1298–1308, M… find our card
  59. Uri Shaham, Elad Segal, Maor Ivgi, Avia Efrat, Ori Yoran, Adi Haviv, Ankit Gupta, Wenhan Xiong, Mor Geva, Jonathan Berant, and Omer Levy. 2022. Scrolls: Standardized comparison over long language sequences. find our card
  60. Zheng Zhao, Shay B Cohen, and Bonnie Webber. 2020. Reducing quantity hallucinations in abstractive summarization. arXiv preprint arXiv:2009.13312. find
  61. James Thorne, Andreas Vlachos, Oana Cocarascu, Christos Christodoulopoulos, and Arpit Mittal. 2018. The fact extraction and VERification (FEVER) shared task. In Proceedings of the First Workshop on Fact Extraction and VERification (FEVER), pages 1–9, Brussels, Belgium. Association for Computational… find
  62. Chunting Zhou, Graham Neubig, Jiatao Gu, Mona Diab, Francisco Guzmán, Luke Zettlemoyer, and Marjan Ghazvininejad. 2021. Detecting hallucinated content in conditional neural sequence generation. In Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, pages 1393– 1404, Online. A… find
  63. 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
  64. Alex Wang, Amanpreet Singh, Julian Michael, Felix Hill, Omer Levy, and Samuel Bowman. 2018. GLUE: A multi-task benchmark and analysis platform for natural language understanding. In Proceedings of the 2018 EMNLP Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP, pages 353–355,… find our card
  65. Yuexiang Xie, Fei Sun, Yang Deng, Yaliang Li, and Bolin Ding. 2021. Factual consistency evaluation for text summarization via counterfactual estimation. In Findings of the Association for Computational Linguistics: EMNLP 2021, pages 100–110, Punta Cana, Dominican Republic. Association for Computatio… find
  66. Wenpeng Yin, Dragomir Radev, and Caiming Xiong. 2021. DocNLI: A large-scale dataset for document-level natural language inference. In Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, pages 4913–4922, Online. Association for Computational Linguistics. find our card
  67. Weizhe Yuan, Graham Neubig, and Pengfei Liu. 2021. BARTScore: Evaluating generated text as text generation. In Advances in Neural Information Processing Systems. find our card
  68. Tianyi Zhang, Varsha Kishore, Felix Wu, Kilian Q. Weinberger, and Yoav Artzi. 2020. Bertscore: Evaluating text generation with bert. In International Conference on Learning Representations. find