Elamur Research
☰
home / papers / 2023 / FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generation

FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generation

Written for the Research library (claude-opus-5-5) from the paper text. Original — aclanthology.org/2023.emnlp-main.741.pdf

Summary

Checking the factuality of long outputs from large language models is hard for two reasons. One response usually mixes correct and incorrect claims, so a single yes/no verdict per sentence or response discards useful information. Verifying every claim by hand is also slow and expensive. The authors propose FActScore, which splits a generation into atomic facts (short statements that each carry one piece of information). It then reports the fraction of those facts that a chosen knowledge source supports, ignoring prompts the model declines to answer. They apply it to people biographies, with Wikipedia as the reference. Expert annotators scored InstructGPT, ChatGPT and the search-augmented PerplexityAI, and all three made many errors, especially for rare entities. To avoid annotation costs, the authors built an automatic estimator that combines retrieval from the knowledge source with a strong LM. Its aggregate scores stay within 2% of the human-derived ones. They then used it to rank a large set of newer models. The work offers a reproducible, fine-grained measure of factual precision and is released as an open package with annotated data.

Key points

  • Human-annotated FActScores were 42% for InstructGPT, 58% for ChatGPT and 71% for PerplexityAI. For ChatGPT, the score falls from 80% on frequent entities to 16% on rare ones.
  • Sentence-level judgments are too coarse. A ChatGPT sentence holds about 4.4 pieces of information on average, and 40% of sentences contain both supported and unsupported content.
  • The automatic estimator has an error rate below 2%. It was used to score 6,500 generations from 13 LMs, which would have cost about $26K with human annotators. The conclusion instead mentions 12 LMs and $65K, so the source text is inconsistent here.
  • GPT-4 and ChatGPT are much more factual than public models but still far below humans. Among public models, Vicuna and Alpaca rank near the top.
  • The tool is available via pip install factscore and on GitHub (shmsw25/FActScore). It measures precision only, not recall, and assumes facts are undebatable, equally weighted and drawn from a non-conflicting source. The authors recommend also reporting the abstention rate and the average number of atomic facts.

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

596 papers of our library cite it (of the 140673 whose bibliographies we read); 20 of them have a page here, 576 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2023.emnlp-main.741 · arXiv 2305.14251

@inproceedings{min2023factscore,
  title = {{FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generation}},
  author = {Min, Sewon and Krishna, Kalpesh and Lyu, Xinxi and Lewis, Mike and Yih, Wen-tau and Koh, Pang and Iyyer, Mohit and Zettlemoyer, Luke and Hajishirzi, Hannaneh},
  booktitle = {Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing},
  year = {2023},
  doi = {10.18653/v1/2023.emnlp-main.741},
  eprint = {2305.14251},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.18653/v1/2023.emnlp-main.741}
}

References

68 references, 24 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Yuvanesh Anand, Zach Nussbaum, Brandon Duder-stadt, Benjamin Schmidt, and Andriy Mulyar. 2023. Gpt4all: Training an assistant-style chatbot with large scale data distillation from gpt-3.5-turbo. https: //github.com/nomic-ai/gpt4all. find
  2. Yuntao Bai, Andy Jones, Kamal Ndousse, Amanda Askell, Anna Chen, Nova DasSarma, Dawn Drain, Stanislav Fort, Deep Ganguli, Tom Henighan, et al. 2022. Training a helpful and harmless assistant with reinforcement learning from human feedback. arXiv preprint arXiv:2204.05862. find
  3. Manik Bhandari, Pranav Narayan Gour, Atabak Ashfaq, Pengfei Liu, and Graham Neubig. 2020. Re-evaluating evaluation in text summarization. In Proceedings of Empirical Methods in Natural Language Processing. find
  4. Stella Biderman, Hailey Schoelkopf, Quentin Anthony, Herbie Bradley, Kyle O’Brien, Eric Hallahan, Mohammad Aflah Khan, Shivanshu Purohit, USVSN Sai Prashanth, Edward Raff, et al. 2023. Pythia: A suite for analyzing large language models across training and scaling. arXiv preprint arXiv:2304.01373. find
  5. Bernd Bohnet, Vinh Tran, Pat Verga, Roee Aharoni, Daniel Andor, Livio Baldini Soares, Massimiliano Ciaramita, Jacob Eisenstein, Kuzman Ganchev, Jonathan Herzig, Kai Hui, Tom Kwiatkowski, Ji Ma, Jianmo Ni, Tal Schuster, Lierni Sestorain Saralegui, William Weston Cohen, Michael Collins, Dipanjan Das,… find our card
  6. Tom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, Sandhini Agarwal, Ariel Herbert-Voss, Gretchen Krueger, Tom Henighan, Rewon Child, Aditya Ramesh, Daniel M. Ziegler, Jeffrey Wu, Clemens Winter,… find
  7. Danqi Chen, Adam Fisch, Jason Weston, and Antoine Bordes. 2017. Reading Wikipedia to answer open-domain questions. In Proceedings of the Association for Computational Linguistics. find our card
  8. Jifan Chen, Aniruddh Sriram, Eunsol Choi, and Greg Durrett. 2022. Generating literal and implied subquestions to fact-check complex claims. In Proceedings of Empirical Methods in Natural Language Processing. find
  9. Sihao Chen, Daniel Khashabi, Wenpeng Yin, Chris Callison-Burch, and Dan Roth. 2019. Seeing things from a different angle:discovering diverse perspectives about claims. In Conference of the North American Chapter of the Association for Computational Linguistics. find
  10. Wei-Lin Chiang, Zhuohan Li, Zi Lin, Ying Sheng, Zhanghao Wu, Hao Zhang, Lianmin Zheng, Siyuan Zhuang, Yonghao Zhuang, Joseph E. Gonzalez, Ion Stoica, and Eric P. Xing. 2023. Vicuna: An open-source chatbot impressing gpt-4 with 90%* chatgpt quality. find
  11. 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. find
  12. Shehzaad Dhuliawala, Mojtaba Komeili, Jing Xu, Roberta Raileanu, Xian Li, Asli Celikyilmaz, and Jason Weston. 2023. Chain-of-verification reduces hallucination in large language models. arXiv preprint arXiv:2309.11495. find our card
  13. Bosheng Ding, Chengwei Qin, Linlin Liu, Lidong Bing, Shafiq Joty, and Boyang Li. 2022. Is gpt-3 a good data annotator? arXiv preprint arXiv:2212.10450. find our card
  14. Alexander Fabbri, Chien-Sheng Wu, Wenhao Liu, and Caiming Xiong. 2022. QAFactEval: Improved QA-based factual consistency evaluation for summarization. In Conference of the North American Chapter of the Association for Computational Linguistics. find our card
  15. Angela Fan, Aleksandra Piktus, Fabio Petroni, Guillaume Wenzek, Marzieh Saeidi, Andreas Vlachos, Antoine Bordes, and Sebastian Riedel. 2020. Generating fact checking briefs. In Proceedings of Empirical Methods in Natural Language Processing. find
  16. Luyu Gao, Zhuyun Dai, Panupong Pasupat, Anthony Chen, Arun Tejasvi Chaganty, Yicheng Fan, Vincent Y Zhao, Ni Lao, Hongrae Lee, Da-Cheng Juan, et al. 2022. Attributed text generation via post-hoc research and revision. arXiv preprint arXiv:2210.08726. find our card
  17. Tianyu Gao, Howard Yen, Jiatong Yu, and Danqi Chen. 2023. Enabling large language models to generate text with citations. find our card
  18. Biyang Guo, Xin Zhang, Ziyuan Wang, Minqi Jiang, Jinran Nie, Yuxuan Ding, Jianwei Yue, and Yupeng Wu. 2023. How close is chatgpt to human experts? comparison corpus, evaluation, and detection. arXiv preprint arxiv:2301.07597. find our card
  19. Kelvin Guu, Tatsunori B. Hashimoto, Yonatan Oren, and Percy Liang. 2018. Generating sentences by editing prototypes. Transactions of the Association for Computational Linguistics. find our card
  20. Jordan Hoffmann, Sebastian Borgeaud, Arthur Mensch, Elena Buchatskaya, Trevor Cai, Eliza Rutherford, Diego de las Casas, Lisa Anne Hendricks, Johannes Welbl, Aidan Clark, Tom Hennigan, Eric Noland, Katherine Millican, George van den Driessche, Bogdan Damoc, Aurelia Guy, Simon Osindero, Karen Simonya… find
  21. Saurav Kadavath, Tom Conerly, Amanda Askell, Tom Henighan, Dawn Drain, Ethan Perez, Nicholas Schiefer, Zac Hatfield Dodds, Nova DasSarma, Eli Tran-Johnson, et al. 2022. Language models (mostly) know what they know. arXiv preprint arXiv:2207.05221. find
  22. Ryo Kamoi, Tanya Goyal, Juan Diego Rodriguez, and Greg Durrett. 2023. Wice: Real-world entailment for claims in wikipedia. arXiv preprint arXiv:2303.01432. find
  23. Nikhil Kandpal, Haikang Deng, Adam Roberts, Eric Wallace, and Colin Raffel. 2022. Large language models struggle to learn long-tail knowledge. arXiv preprint arXiv:2211.08411. find our card
  24. Kalpesh Krishna, Erin Bransom, Bailey Kuehl, Mohit Iyyer, Pradeep Dasigi, Arman Cohan, and Kyle Lo. 2023. LongEval: Guidelines for human evaluation of faithfulness in long-form summarization. In Proceedings of the European Chapter of the Association for Computational Linguistics. find
  25. Wojciech Kryscinski, Bryan McCann, Caiming Xiong, and Richard Socher. 2020. Evaluating the factual consistency of abstractive text summarization. In Proceedings of Empirical Methods in Natural Language Processing. find
  26. Philippe Laban, Tobias Schnabel, Paul N. Bennett, and Marti A. Hearst. 2022. SummaC: Re-visiting NLI-based models for inconsistency detection in summarization. Transactions of the Association for Computational Linguistics. find
  27. Nayeon Lee, Wei Ping, Peng Xu, Mostofa Patwary, Pascale Fung, Mohammad Shoeybi, and Bryan Catanzaro. 2022. Factuality enhanced language models for open-ended text generation. In Advances in Neural Information Processing Systems. find
  28. Peter Lee, Sebastien Bubeck, and Joseph Petro. 2023. Benefits, limits, and risks of gpt-4 as an ai chatbot for medicine. New England Journal of Medicine. find
  29. Jimmy Lin, Xueguang Ma, Sheng-Chieh Lin, Jheng-Hong Yang, Ronak Pradeep, and Rodrigo Nogueira. 2021. Pyserini: A Python toolkit for reproducible information retrieval research with sparse and dense representations. In Proceedings of the ACM SIGIR Conference on Research and Development in Information… find
  30. Nelson F Liu, Tianyi Zhang, and Percy Liang. 2023a. Evaluating verifiability in generative search engines. arXiv preprint arXiv:2304.09848. find our card
  31. Yang Liu, Dan Iter, Yichong Xu, Shuohang Wang, Ruochen Xu, and Chenguang Zhu. 2023b. Gpteval: Nlg evaluation using gpt-4 with better human alignment. arXiv preprint arXiv:2303.16634. find
  32. Yixin Liu, Alexander R Fabbri, Pengfei Liu, Yilun Zhao, Linyong Nan, Ruilin Han, Simeng Han, Shafiq Joty, Chien-Sheng Wu, Caiming Xiong, et al. 2022. Revisiting the gold standard: Grounding summarization evaluation with robust human evaluation. arXiv preprint arXiv:2212.07981. find our card
  33. Qingsong Ma, Johnny Wei, Ondřej Bojar, and Yvette Graham. 2019. Results of the WMT19 metrics shared task: Segment-level and strong MT systems pose big challenges. In Proceedings of the Fourth Conference on Machine Translation. find our card
  34. Potsawee Manakul, Adian Liusie, and Mark JF Gales. 2023. Selfcheckgpt: Zero-resource black-box hallucination detection for generative large language models. arXiv preprint arXiv:2303.08896. find
  35. Tsvetomila Mihaylova, Georgi Karadzhov, Pepa Atanasova, Ramy Baly, Mitra Mohtarami, and Preslav Nakov. 2019. SemEval-2019 task 8: Fact checking in community question answering forums. In Proceedings of the 13th International Workshop on Semantic Evaluation. find
  36. Sewon Min, Weijia Shi, Mike Lewis, Xilun Chen, Wen-tau Yih, Hannaneh Hajishirzi, and Luke Zettlemoyer. 2023. Nonparametric masked language modeling. In Findings of the Association for Computational Linguistics: ACL. find
  37. Preslav Nakov, Alberto Barrón-Cedeno, Tamer Elsayed, Reem Suwaileh, Lluís Màrquez, Wajdi Zaghouani, Pepa Atanasova, Spas Kyuchukov, and Giovanni Da San Martino. 2018. Overview of the clef-2018 checkthat! lab on automatic identification and verification of political claims. In Experimental IR Meets M… find
  38. Ani Nenkova and Rebecca Passonneau. 2004. Evaluating content selection in summarization: The pyramid method. In Conference of the North American Chapter of the Association for Computational Linguistics. find
  39. Jianmo Ni, Chen Qu, Jing Lu, Zhuyun Dai, Gustavo Hernandez Abrego, Ji Ma, Vincent Zhao, Yi Luan, Keith Hall, Ming-Wei Chang, and Yinfei Yang. 2022. Large dual encoders are generalizable retrievers. In Proceedings of Empirical Methods in Natural Language Processing. find our card
  40. Harsha Nori, Nicholas King, Scott Mayer McKinney, Dean Carignan, and Eric Horvitz. 2023. Capabilities of gpt-4 on medical challenge problems. arXiv preprint arXiv:2303.13375. find our card
  41. OpenAI. 2022. Chatgpt blog post. https://openai. com/blog/chatgpt. find
  42. OpenAI. 2023. Gpt-4 technical report. arXiv preprint arXiv:2303.08774. find our card
  43. Chaitanya Malaviya, Subin Lee, Sihao Chen, Elizabeth Sieber, Mark Yatskar, and Dan Roth. 2023. Expertqa: Expert-curated questions and attributed answers. arXiv preprint arXiv:2309.07852. find
  44. Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Gray, John Schulman, Jacob Hilton, Fraser Kelton, Luke Miller, Maddie Simens, Amanda Askell, Peter Welinder, Paul Christiano, Jan Leike, and Ryan Lowe. 2022. Trai… find
  45. Alex Mallen, Akari Asai, Victor Zhong, Rajarshi Das, Daniel Khashabi, and Hannaneh Hajishirzi. 2023. When not to trust language models: Investigating effectiveness of parametric and non-parametric memories. In Proceedings of the Association for Computational Linguistics. find
  46. Artidoro Pagnoni, Vidhisha Balachandran, and Yulia Tsvetkov. 2021. Understanding factuality in abstractive summarization with FRANK: A benchmark for factuality metrics. In Conference of the North American Chapter of the Association for Computational Linguistics. find our card
  47. Fabio Petroni, Aleksandra Piktus, Angela Fan, Patrick Lewis, Majid Yazdani, Nicola De Cao, James Thorne, Yacine Jernite, Vladimir Karpukhin, Jean Maillard, Vassilis Plachouras, Tim Rocktäschel, and Sebastian Riedel. 2021. KILT: a benchmark for knowledge intensive language tasks. In Conference of the… find our card
  48. Pranav Rajpurkar, Jian Zhang, Konstantin Lopyrev, and Percy Liang. 2016. SQuAD: 100,000+ questions for machine comprehension of text. In Proceedings of Empirical Methods in Natural Language Processing. find our card
  49. Hannah Rashkin, Vitaly Nikolaev, Matthew Lamm, Lora Aroyo, Michael Collins, Dipanjan Das, Slav Petrov, Gaurav Singh Tomar, Iulia Turc, and David Reitter. 2021. Measuring attribution in natural language generation models. arXiv preprint arXiv:2112.12870. find our card
  50. Shaden Shaar, Firoj Alam, Giovanni Da San Martino, and Preslav Nakov. 2022. The role of context in detecting previously fact-checked claims. In Findings of the Association for Computational Linguistics: NAACL 2022. find
  51. Ori Shapira, David Gabay, Yang Gao, Hadar Ronen, Ramakanth Pasunuru, Mohit Bansal, Yael Amsterdamer, and Ido Dagan. 2019. Crowdsourcing lightweight pyramids for manual summary evaluation. In Conference of the North American Chapter of the Association for Computational Linguistics. find
  52. Kurt Shuster, Spencer Poff, Moya Chen, Douwe Kiela, and Jason Weston. 2021. Retrieval augmentation reduces hallucination in conversation. In Findings of the Association for Computational Linguistics: EMNLP 2021. find our card
  53. Simeng Sun, Dhawal Gupta, and Mohit Iyyer. 2023. Exploring the impact of low-rank adaptation on the performance, efficiency, and regularization of rlhf. arXiv preprint arXiv:2309.09055. find
  54. Rohan Taori, Ishaan Gulrajani, Tianyi Zhang, Yann Dubois, Xuechen Li, Carlos Guestrin, Percy Liang, and Tatsunori B. Hashimoto. 2023. Stanford alpaca: An instruction-following llama model. https:// github.com/tatsu-lab/stanford\_alpaca. find
  55. Brian Thompson and Matt Post. 2020. Automatic machine translation evaluation in many languages via zero-shot paraphrasing. In Proceedings of Empirical Methods in Natural Language Processing. find our card
  56. James Thorne, Andreas Vlachos, Christos Christodoulopoulos, and Arpit Mittal. 2018. FEVER: a large-scale dataset for fact extraction and VERification. In Conference of the North American Chapter of the Association for Computational Linguistics. find our card
  57. Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Faisal Azhar, et al. 2023. Llama: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971. find
  58. David Wadden, Shanchuan Lin, Kyle Lo, Lucy Lu Wang, Madeleine van Zuylen, Arman Cohan, and Hannaneh Hajishirzi. 2020. Fact or fiction: Verifying scientific claims. In Proceedings of Empirical Methods in Natural Language Processing. find our card
  59. David Wadden, Kyle Lo, Bailey Kuehl, Arman Cohan, Iz Beltagy, Lucy Lu Wang, and Hannaneh Hajishirzi. 2022. SciFact-open: Towards open-domain scientific claim verification. In Findings of the Association for Computational Linguistics: EMNLP. find
  60. Alex Wang, Kyunghyun Cho, and Mike Lewis. 2020. Asking and answering questions to evaluate the factual consistency of summaries. In Proceedings of the Association for Computational Linguistics. find
  61. Yizhong Wang, Swaroop Mishra, Pegah Alipoormolabashi, Yeganeh Kordi, Amirreza Mirzaei, Atharva Naik, Arjun Ashok, Arut Selvan Dhanasekaran, Anjana Arunkumar, David Stap, Eshaan Pathak, Giannis Karamanolakis, Haizhi Lai, Ishan Purohit, Ishani Mondal, Jacob Anderson, Kirby Kuznia, Krima Doshi, Kuntal… find
  62. John Wieting, Kevin Gimpel, Graham Neubig, and Taylor Berg-kirkpatrick. 2022. Paraphrastic representations at scale. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing: System Demonstrations. find
  63. Dustin Wright, David Wadden, Kyle Lo, Bailey Kuehl, Arman Cohan, Isabelle Augenstein, and Lucy Lu Wang. 2022. Generating scientific claims for zero-shot scientific fact checking. In Proceedings of the Association for Computational Linguistics. find
  64. Fangyuan Xu, Yixiao Song, Mohit Iyyer, and Eunsol Choi. 2023. A critical evaluation of evaluations for long-form question answering. In Proceedings of the Association for Computational Linguistics. find
  65. Seonghyeon Ye, Doyoung Kim, Sungdong Kim, Hyeonbin Hwang, Seungone Kim, Yongrae Jo, James Thorne, Juho Kim, and Minjoon Seo. 2023. Flask: Fine-grained language model evaluation based on alignment skill sets. arXiv preprint arXiv:2307.10928. find
  66. Xiang Yue, Boshi Wang, Kai Zhang, Ziru Chen, Yu Su, and Huan Sun. 2023. Automatic evaluation of attribution by large language models. arXiv preprint arXiv:2305.06311. find
  67. Shiyue Zhang and Mohit Bansal. 2021. Finding a balanced degree of automation for summary evaluation. In Proceedings of Empirical Methods in Natural Language Processing. find
  68. Tianyi Zhang, Varsha Kishore, Felix Wu, Kilian Q. Weinberger, and Yoav Artzi. 2020. Bertscore: Evaluating text generation with bert. In Proceedings of the International Conference on Learning Representations. find