Elamur Research⌕
☰
home / papers / 2024 / ExpertQA: Expert-Curated Questions and Attributed Answers

ExpertQA: Expert-Curated Questions and Attributed Answers

Summary

LLMs are increasingly used by specialists, but it is unclear how factual their answers are and how well their citations hold up in specific professional fields, especially high-risk ones like medicine and law. To study this, the authors recruited qualified experts via Prolific. Each expert needed formal training and at least three years of work experience. The experts wrote difficult questions from their own practice, which yielded 2177 questions across 32 fields. Six attribution-producing systems answered these questions: purely generative, retrieve-and-read, and post-hoc attribution setups. Each answer was split into sentence-level claims. The same experts then rated their own answers' claims for informativeness, factuality, citeworthiness, evidence support and source reliability, and rewrote claims to be correct and properly sourced.

Retrieve-and-read systems cited more completely but still missed many claims that needed citations. The choice of retrieval source strongly affected both attribution quality and factuality. Off-the-shelf automatic attribution and factuality metrics agreed poorly with expert judgments. The resulting ExpertQA benchmark supports work on attribution, factuality estimation and long-form QA.

Key points

  • ExpertQA contains 2177 expert-written questions from 32 fields. It includes claim-level expert judgments and expert-revised answers with verified attributions, available on GitHub (chaitanyamalaviya/ExpertQA).
  • The paper reports the number of contributing experts inconsistently: 484 in the abstract and acknowledgements, but 524 in Section 2.1.
  • In medicine, 35% of attributions were incomplete; in law, 31% were. Overall, 51% of attributions were not rated reliable by experts.
  • Automatic estimators, such as the attribution method of Bohnet et al. (2022) and the factuality method of Min et al. (2023), correlated weakly with expert labels. Finetuning them on ExpertQA improved results across domains.
  • Long-form QA baselines finetuned on ExpertQA improved on ROUGE and QAFactEval, but substantial headroom remains.

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

Institutions

🎓University of WashingtonSeattle · United States206 works · 158,423 citations🎓University of PennsylvaniaPhiladelphia · United States143 works · 56,645 citations

Cite

Cite as Chaitanya Malaviya, Subin Lee, Sihao Chen, Elizabeth Sieber, Mark Yatskar and Dan Roth. 2024. ExpertQA: Expert-Curated Questions and Attributed Answers. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2024.naacl-long.167

@inproceedings{malaviya2024expertqa,
  title = {{ExpertQA: Expert-Curated Questions and Attributed Answers}},
  author = {Chaitanya Malaviya, Subin Lee, Sihao Chen, Elizabeth Sieber, Mark Yatskar and Dan Roth},
  booktitle = {Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)},
  year = {2024},
  doi = {10.18653/v1/2024.naacl-long.167},
  url = {https://doi.org/10.18653/v1/2024.naacl-long.167}
}

References

74 references, 32 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Ayush Agrawal, Lester Mackey, and Adam Tauman Kalai. 2023. Do language models know when they’re hallucinating references? arXiv preprint arXiv:2305.18248. find
  2. Pepa Atanasova, Jakob Grue Simonsen, Christina Lioma, and Isabelle Augenstein. 2022. Fact checking with insufficient evidence. Transactions of the Association for Computational Linguistics, 10:746–763. find
  3. Isabelle Augenstein, Timothy Baldwin, Meeyoung Cha, Tanmoy Chakraborty, Giovanni Luca Ciampaglia, David Corney, Renee DiResta, Emilio Ferrara, Scott Hale, Alon Halevy, et al. 2023. Factuality challenges in the era of large language models. arXiv preprint arXiv:2310.05189. our card
  4. Amos Azaria and Tom Mitchell. 2023. The internal state of an llm knows when its lying. arXiv preprint arXiv:2304.13734. our card
  5. Abeba Birhane, Atoosa Kasirzadeh, David Leslie, and Sandra Wachter. 2023. Science in the age of large language models. Nature Reviews Physics, 5(5):277– 280. find
  6. Bernd Bohnet, Vinh Q Tran, Pat Verga, Roee Aharoni, Daniel Andor, Livio Baldini Soares, Jacob Eisenstein, Kuzman Ganchev, Jonathan Herzig, Kai Hui, et al. 2022. Attributed question answering: Evaluation and modeling for attributed large language models. arXiv preprint arXiv:2212.08037. our card
  7. Sebastian Borgeaud, Arthur Mensch, Jordan Hoffmann, Trevor Cai, Eliza Rutherford, Katie Millican, George Bm Van Den Driessche, Jean-Baptiste Lespiau, Bogdan Damoc, Aidan Clark, et al. 2022. Improving language models by retrieving from trillions of tokens. In International conference on machine learn… find
  8. Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared D Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, et al. 2020. Language models are few-shot learners. Advances in neural information processing systems, 33:1877–1901. our card
  9. Danqi Chen, Adam Fisch, Jason Weston, and Antoine Bordes. 2017. Reading Wikipedia to answer open-domain questions. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1870–1879, Vancouver, Canada. Association for Computational Lin… our card
  10. Jifan Chen, Grace Kim, Aniruddh Sriram, Greg Durrett, and Eunsol Choi. 2023. Complex claim verification with evidence retrieved in the wild. arXiv preprint arXiv:2305.11859. find
  11. Sihao Chen, Senaka Buthpitiya, Alex Fabrikant, Dan Roth, and Tal Schuster. 2022. Propsegment: A large-scale corpus for proposition-level segmentation and entailment recognition. arXiv preprint arXiv:2212.10750. find
  12. 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
  13. Hyung Won Chung, Le Hou, Shayne Longpre, Barret Zoph, Yi Tay, William Fedus, Eric Li, Xuezhi Wang, Mostafa Dehghani, Siddhartha Brahma, et al. 2022. Scaling instruction-finetuned language models. arXiv preprint arXiv:2210.11416. our card
  14. Debadutta Dash, Rahul Thapa, Juan M Banda, Akshay Swaminathan, Morgan Cheatham, Mehr Kashyap, Nikesh Kotecha, Jonathan H Chen, Saurabh Gombar, Lance Downing, et al. 2023. Evaluation of gpt-3.5 and gpt-4 for supporting real-world information needs in healthcare delivery. arXiv preprint arXiv:2304.137… find
  15. Cicero Dos Santos, Luciano Barbosa, Dasha Bogdanova, and Bianca Zadrozny. 2015. Learning hybrid representations to retrieve semantically equivalent questions. In Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natu… find
  16. Nouha Dziri, Hannah Rashkin, Tal Linzen, and David Reitter. 2022. Evaluating attribution in dialogue systems: The begin benchmark. Transactions of the Association for Computational Linguistics, 10:1066– 1083. find
  17. Owain Evans, Owen Cotton-Barratt, Lukas Finnveden, Adam Bales, Avital Balwit, Peter Wills, Luca Righetti, and William Saunders. 2021. Truthful ai: Developing and governing ai that does not lie. arXiv preprint arXiv:2110.06674. — [arXiv:2110.06674](https://arxiv.org/abs/2110.06674) find
  18. Alexander R. Fabbri, Chien-Sheng Wu, Wenhao Liu, and Caiming Xiong. 2022. Qafacteval: Improved qa-based factual consistency evaluation for summarization. our card
  19. Angela Fan, Yacine Jernite, Ethan Perez, David Grangier, Jason Weston, and Michael Auli. 2019. Eli5: Long form question answering. arXiv preprint arXiv:1907.09190. our card
  20. Shangbin Feng, Vidhisha Balachandran, Yuyang Bai, and Yulia Tsvetkov. 2023. Factkb: Generalizable factuality evaluation using language models enhanced with factual knowledge. arXiv preprint arXiv:2305.08281. find
  21. Luyu Gao, Zhuyun Dai, Panupong Pasupat, Anthony Chen, Arun Tejasvi Chaganty, Yicheng Fan, Vincent Zhao, Ni Lao, Hongrae Lee, Da-Cheng Juan, and Kelvin Guu. 2023. RARR: Researching and revising what language models say, using language models. In Proceedings of the 61st Annual Meeting of the Associati… our card
  22. Neel Guha, Julian Nyarko, Daniel E Ho, Christopher Ré, Adam Chilton, Aditya Narayana, Alex Chohlas-Wood, Austin Peters, Brandon Waldon, Daniel N Rockmore, et al. 2023. Legalbench: A collaboratively built benchmark for measuring legal reasoning in large language models. arXiv preprint arXiv:2308.1146… our card
  23. 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
  24. Ehsan Kamalloo, Aref Jafari, Xinyu Zhang, Nandan Thakur, and Jimmy Lin. 2023. Hagrid: A human-llm collaborative dataset for generative information-seeking with attribution. arXiv preprint arXiv:2307.16883. — [arXiv:2307.16883](https://arxiv.org/abs/2307.16883) find
  25. Kelvin Guu, Kenton Lee, Zora Tung, Panupong Pasupat, and Mingwei Chang. 2020. Retrieval augmented language model pre-training. In International conference on machine learning, pages 3929–3938. PMLR. find
  26. 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
  27. Hangfeng He, Hongming Zhang, and Dan Roth. 2022. Rethinking with retrieval: Faithful large language model inference. arXiv preprint arXiv:2301.00303. find
  28. Mario Krenn, Robert Pollice, Si Yue Guo, Matteo Aldeghi, Alba Cervera-Lierta, Pascal Friederich, Gabriel dos Passos Gomes, Florian Häse, Adrian Jinich, AkshatKumar Nigam, Zhenpeng Yao, and Alán Aspuru-Guzik. 2022. On scientific understanding with artificial intelligence. Nature Reviews Physics, 4(12… find
  29. Dan Hendrycks, Collin Burns, Steven Basart, Andy Zou, Mantas Mazeika, Dawn Song, and Jacob Steinhardt. 2021. Measuring massive multitask language understanding. Proceedings of the International Conference on Learning Representations (ICLR). our card
  30. Matthew Honnibal and Ines Montani. 2017. spaCy 2: Natural language understanding with Bloom embeddings, convolutional neural networks and incremental parsing. To appear. find
  31. Or Honovich, Roee Aharoni, Jonathan Herzig, Hagai Taitelbaum, Doron Kukliansy, Vered Cohen, Thomas Scialom, Idan Szpektor, Avinatan Hassidim, and Yossi Matias. 2022. TRUE: Re-evaluating factual consistency evaluation. In Proceedings of the 2022 Conference of the North American Chapter of the Associa… our card
  32. Gautier Izacard, Patrick Lewis, Maria Lomeli, Lucas Hosseini, Fabio Petroni, Timo Schick, Jane Dwivedi-Yu, Armand Joulin, Sebastian Riedel, and Edouard Grave. 2022. Few-shot learning with retrieval augmented language models. arXiv preprint arXiv:2208.03299. find
  33. Di Jin, Eileen Pan, Nassim Oufattole, Wei-Hung Weng, Hanyi Fang, and Peter Szolovits. 2021. What disease does this patient have? a large-scale open domain question answering dataset from medical exams. Applied Sciences, 11(14):6421. our card
  34. Qiao Jin, Bhuwan Dhingra, Zhengping Liu, William Cohen, and Xinghua Lu. 2019. PubMedQA: A dataset for biomedical research question answering. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Proce… our card
  35. Kalpesh Krishna, Aurko Roy, and Mohit Iyyer. 2021. Hurdles to progress in long-form question answering. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 4940–4957, Online. Association for Computat… our card
  36. 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
  37. 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, 388(13):1233–1239. find
  38. 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. our card
  39. Stephanie Lin, Jacob Hilton, and Owain Evans. 2021. Truthfulqa: Measuring how models mimic human falsehoods. arXiv preprint arXiv:2109.07958. our card
  40. Nelson F Liu, Tianyi Zhang, and Percy Liang. 2023. Evaluating verifiability in generative search engines. arXiv preprint arXiv:2304.09848. our card
  41. 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
  42. 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
  43. Jacob Menick, Maja Trebacz, Vladimir Mikulik, John Aslanides, Francis Song, Martin Chadwick, Mia Glaese, Susannah Young, Lucy Campbell-Gillingham, Geoffrey Irving, et al. 2022. Teaching language models to support answers with verified quotes. arXiv preprint arXiv:2203.11147. our card
  44. Donald Metzler, Yi Tay, Dara Bahri, and Marc Najork. 2021. Rethinking search: making domain experts out of dilettantes. In Acm sigir forum, volume 55, pages 1–27. ACM New York, NY, USA. find
  45. Sewon Min, Kalpesh Krishna, Xinxi Lyu, Mike Lewis, Wen-tau Yih, Pang Wei Koh, Mohit Iyyer, Luke Zettlemoyer, and Hannaneh Hajishirzi. 2023. Factscore: Fine-grained atomic evaluation of factual precision in long form text generation. arXiv preprint arXiv:2305.14251v1. our card
  46. Abhika Mishra, Akari Asai, Vidhisha Balachandran, Yizhong Wang, Graham Neubig, Yulia Tsvetkov, and Hannaneh Hajishirzi. 2024. Fine-grained hallucinations detections. arXiv preprint. find
  47. Dor Muhlgay, Ori Ram, Inbal Magar, Yoav Levine, Nir Ratner, Yonatan Belinkov, Omri Abend, Kevin Leyton-Brown, Amnon Shashua, and Yoav Shoham. 2023. Generating benchmarks for factuality evaluation of language models. arXiv preprint arXiv:2307.06908. find
  48. Benjamin Muller, John Wieting, Jonathan H Clark, Tom Kwiatkowski, Sebastian Ruder, Livio Baldini Soares, Roee Aharoni, Jonathan Herzig, and Xinyi Wang. 2023. Evaluating and modeling attribution for cross-lingual question answering. arXiv preprint arXiv:2305.14332. find
  49. Reiichiro Nakano, Jacob Hilton, Suchir Balaji, Jeff Wu, Long Ouyang, Christina Kim, Christopher Hesse, Shantanu Jain, Vineet Kosaraju, William Saunders, et al. 2021. Webgpt: Browser-assisted question-answering with human feedback. arXiv preprint arXiv:2112.09332. our card
  50. Tri Nguyen, Mir Rosenberg, Xia Song, Jianfeng Gao, Saurabh Tiwary, Rangan Majumder, and Li Deng. 2016. Ms marco: A human generated machine reading comprehension dataset. choice, 2640:660. our card
  51. OpenAI. 2023. Gpt-4 technical report. abs/2303.08774. our card
  52. ArXiv, Brian Owens. 2023. How nature readers are using chatgpt. Nature. find
  53. 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… our card
  54. Ankit Pal, Logesh Kumar Umapathi, and Malaikannan Sankarasubbu. 2022. Medmcqa: A large-scale multi-subject multi-choice dataset for medical domain question answering. In Proceedings of the Conference on Health, Inference, and Learning, volume 174 of Proceedings of Machine Learning Research, pages 24… our card
  55. Anusri Pampari, Preethi Raghavan, Jennifer Liang, and Jian Peng. 2018. emrQA: A large corpus for question answering on electronic medical records. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 2357–2368, Brussels, Belgium. Association for Computatio… our card
  56. Denis Peskoff and Brandon Stewart. 2023. Credible without credit: Domain experts assess generative language models. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 427–438, Toronto, Canada. Association for Computational Lingu… find
  57. Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Dmytro Okhonko, Samuel Broscheit, Gautier Izacard, Patrick Lewis, Barlas Oguz, Edouard Grave, Wen-tau Yih, and Sebastian Riedel. 2021. The web is your oyster - knowledge-intensive NLP against a very large web corpus. CoRR, abs/2112.09924. — [arXiv:2112.09924](https://arxiv.org/abs/2112.09924) find
  58. Samyam Rajbhandari, Jeff Rasley, Olatunji Ruwase, and Yuxiong He. 2020. Zero: Memory optimizations toward training trillion parameter models. In SC20: International Conference for High Performance Computing, Networking, Storage and Analysis, pages 1– 16. IEEE. find
  59. 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. our card
  60. Siva Reddy, Danqi Chen, and Christopher D. Manning. 2019. CoQA: A conversational question answering challenge. Transactions of the Association for Computational Linguistics, 7:249–266. our card
  61. Stephen Robertson, Hugo Zaragoza, et al. 2009. The probabilistic relevance framework: Bm25 and beyond. Foundations and Trends® in Information Retrieval, 3(4):333–389. find
  62. Anna Rogers, Olga Kovaleva, Matthew Downey, and Anna Rumshisky. 2020. Getting closer to ai complete question answering: A set of prerequisite real tasks. In Proceedings of the AAAI conference on artificial intelligence, volume 34, pages 8722–8731. find
  63. Daniel E Rose and Danny Levinson. 2004. Understanding user goals in web search. In Proceedings of the 13th international conference on World Wide Web, pages 13–19. find
  64. Tal Schuster, Sihao Chen, Senaka Buthpitiya, Alex Fabrikant, and Donald Metzler. 2022. Stretching sentence-pair NLI models to reason over long documents and clusters. In Findings of the Association for Computational Linguistics: EMNLP 2022, pages 394–412, Abu Dhabi, United Arab Emirates. Association… find
  65. Eugene Volokh. 2023. Large libel models? liability for ai output. find
  66. Fangyuan Xu, Yixiao Song, Mohit Iyyer, and Eunsol Choi. 2023. A critical evaluation of evaluations for long-form question answering. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 3225–3245, Toronto, Canada. Association for C… find
  67. Ivan Stelmakh, Yi Luan, Bhuwan Dhingra, and Ming-Wei Chang. 2022. Asqa: Factoid questions meet long-form answers. arXiv preprint arXiv:2204.06092. our card
  68. Liyan Tang, Igor Shalyminov, Amy Wing mei Wong, Jon Burnsky, Jake W. Vincent, Yu’an Yang, Siffi Singh, Song Feng, Hwanjun Song, Hang Su, Lijia Sun, Yi Zhang, Saab Mansour, and Kathleen McKeown. 2024. Tofueval: Evaluating hallucinations of llms on topic-focused dialogue summarization. find
  69. 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. ↗
  70. Yi Tay, Vinh Tran, Mostafa Dehghani, Jianmo Ni, Dara Bahri, Harsh Mehta, Zhen Qin, Kai Hui, Zhe Zhao, Jai Gupta, et al. 2022. Transformer memory as a differentiable search index. Advances in Neural Information Processing Systems, 35:21831–21843. our card
  71. Romal Thoppilan, Daniel De Freitas, Jamie Hall, Noam Shazeer, Apoorv Kulshreshtha, Heng-Tze Cheng, Alicia Jin, Taylor Bos, Leslie Baker, Yu Du, et al. 2022. Lamda: Language models for dialog applications. arXiv preprint arXiv:2201.08239. our card
  72. 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… our card
  73. Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi, Yasmine Babaei, Nikolay Bashlykov, Soumya Batra, Prajjwal Bhargava, Shruti Bhosale, et al. 2023. Llama 2: Open foundation and fine-tuned chat models. arXiv preprint arXiv:2307.09288. find
  74. George Tsatsaronis, Georgios Balikas, Prodromos Malakasiotis, Ioannis Partalas, Matthias Zschunke, Michael R Alvers, Dirk Weissenborn, Anastasia Krithara, Sergios Petridis, Dimitris Polychronopoulos, et al. 2015. An overview of the bioasq large-scale biomedical semantic indexing and question answeri… find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/malaviya-2024-expertqa-expert-curated-questions-attributed/