ChatGPT Hallucinates when Attributing Answers
Open accessThe DOI matches its Crossref record; 12 of the 12 references it cites were found in a registry or an index.
Abstract
Can ChatGPT provide evidence to support its answers? Does the evidence it suggests actually exist and does it really support the answer? We investigate these questions using a collection of domain-specific knowledge-based questions, specifically prompting ChatGPT to provide both an answer and supporting evidence in the form of references to external sources. We also investigate how different prompts impact answers and evidence.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Survey of Hallucination in Natural Language Generation
Ji, Z., Lee, N., Frieske, R., Yu, T., Su, D., Xu, Y., Ishii, E., Bang, Y. J., Madotto, A., & Fung, P. (2023) · ACM Computing Surveys · doi:10.1145/3571730 · Cited by 5,025 · cited 1×
“Hallucinations in LLMs like ChatGPT refer to the generation of factually incorrect or entirely made-up information [9].”
BackgroundSallam, M. (2023) · Healthcare · doi:10.3390/healthcare11060887 · Cited by 3,126 · cited 1×
“The recent progress in instruction-based large language models (LLMs), for instance, ChatGPT, has demonstrated their proficiency in adhering to user guidelines to successfully accomplish tasks [4, 6, 18, 22].”
BackgroundLaMDA: Language Models for Dialog Applications
Thoppilan, R., De Freitas, D., Hall, J., Shazeer, N., Kulshreshtha, A., Cheng, H.-T., Jin, A., Bos, T., Baker, L., Du, Y., et al. (2022) · arXiv · doi:10.48550/arxiv.2201.08239 · Cited by 778 · cited 2×
“Recent research has investigated techniques for attribution [1, 2, 12, 13, 17, 21]; but the validity of the attributions produced by the popular ChatGPT model, which at the time of writing boasts a user base in excess of 100 million people, has received little attention.”
BackgroundTaxonomy of Risks posed by Language Models
Weidinger, L., Uesato, J., Rauh, M., Griffin, C., Huang, P.-S., Mellor, J., Glaese, A., Cheng, M., Balle, B., Kasirzadeh, A., et al. (2022) · 2022 ACM Conference on Fairness Accountability and Transparency · doi:10.1145/3531146.3533088 · Cited by 764 · cited 2×
“The reliance on retrieve-then-generate pipelines, where a generation occurs from evidence retrieved through an initial round of search, and the use of attribution have been poised to be possible mitigation strategies [23].”
BackgroundCan ChatGPT Write a Good Boolean Query for Systematic Review Literature Search?
Wang, S., Scells, H., Koopman, B., & Zuccon, G. (2023) · Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval · doi:10.1145/3539618.3591703 · Cited by 275 · cited 1×
“The recent progress in instruction-based large language models (LLMs), for instance, ChatGPT, has demonstrated their proficiency in adhering to user guidelines to successfully accomplish tasks [4, 6, 18, 22].”
BackgroundSelf-citationIs ChatGPT Good at Search? Investigating Large Language Models as Re-Ranking Agents
Sun, W., Yan, L., Ma, X., Wang, S., Ren, P., Chen, Z., Yin, D., & Ren, Z. (2023) · arXiv · doi:10.48550/arxiv.2304.09542 · Cited by 270
ChatGPT is not all you need. A state-of-the-art review of large generative AI models
Gozalo-Brizuela, R., & Garrido-Merchan, E. C. (2023) · arXiv · doi:10.48550/arxiv.2301.04655 · Cited by 245
Gravel, J., D’Amours-Gravel, M., & Osmanlliu, E. (2023) · Mayo Clinic Proceedings: Digital Health · doi:10.1016/j.mcpdig.2023.05.004 · Cited by 222 · cited 1×
“We notice that while the quality and correctness of citations and references produced by LLMs have been somewhat investigated in the context of specific techniques aimed at offering attribution capabilities [1, 3, 12, 13], the evaluation of attribution references produced by the popular ChatGPT has been limited [5, 7].”
BackgroundGravel, J., D’Amours-Gravel, M., & Osmanlliu, E. (2023) · doi:10.1101/2023.03.16.23286914 · Cited by 220
Rethinking search: making domain experts out of dilettantes
Metzler, D., Tay, Y., Bahri, D., & Najork, M. (2021) · ACM SIGIR Forum · doi:10.1145/3476415.3476428 · Cited by 112
Research works printed but not linked: 12
- Bernd Bohnet, Vinh Q Tran, Pat Verga, Roee Aharoni, Daniel Andor, Livio Baldini Soares, Jacob Eisenstein, Kuzman Ganchev, Jonathan Herzig, Kai Hui, 2022. Attributed Question Answering: Evaluation and Modeling for Attributed Large Language Models. arXiv preprint arXiv:2212.08037 (2022). · arXiv
- Tianyu Gao, Howard Yen, Jiatong Yu, and Danqi Chen. 2023. Enabling Large Language Models to Generate Text with Citations. arXiv preprint arXiv:2305.14627 (2023). · arXiv
- 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 (2023). · arXiv
- Yunjie Ji, Yan Gong, Yiping Peng, Chao Ni, Peiyan Sun, Dongyu Pan, Baochang Ma, and Xiangang Li. 2023. Exploring ChatGPT’s Ability to Rank Content: A Preliminary Study on Consistency with Human Preferences. arXiv preprint arXiv:2303.07610 (2023). · arXiv
- Bevan Koopman, Ahmed Mourad, Hang Li, Anton van der Vegt, Shengyao Zhuang, Simon Gibson, Yash Dang, David Lawrence, and Guido Zuccon. 2023. AgAsk: an agent to help answer farmer’s questions from scientific documents. International Journal on Digital Libraries (2023), 1–16.
- Stephanie Lin, Jacob Hilton, and Owain Evans. 2021. Truthfulqa: Measuring how models mimic human falsehoods. arXiv preprint arXiv:2109.07958 (2021). · arXiv
- Jacob Menick, Maja Trebacz, Vladimir Mikulik, John Aslanides, Francis Song, Martin Chadwick, Mia Glaese, Susannah Young, Lucy Campbell-Gillingham, Geoffrey Irving, 2022. Teaching language models to support answers with verified quotes. arXiv preprint arXiv:2203.11147 (2022). · arXiv
- Reham Omar, Omij Mangukiya, Panos Kalnis, and Essam Mansour. 2023. Chatgpt versus traditional question answering for knowledge graphs: Current status and future directions towards knowledge graph chatbots. arXiv preprint arXiv:2302.06466 (2023). · arXiv
- Jack W Rae, Sebastian Borgeaud, Trevor Cai, Katie Millican, Jordan Hoffmann, Francis Song, John Aslanides, Sarah Henderson, Roman Ring, Susannah Young, 2021. Scaling language models: Methods, analysis & insights from training gopher. arXiv preprint arXiv:2112.11446 (2021). · arXiv
- Hannah Rashkin Vitaly Nikolaev Matthew Lamm Lora Aroyo Michael Collins Dipanjan Das Slav Petrov Gaurav Singh Tomar Iulia Turc and David Reitter. [n.d.]. Measuring attribution in natural language generation models. Computational Linguistics ([n. d.]) 1–66. · DOI
- Yiming Tan, Dehai Min, Yu Li, Wenbo Li, Nan Hu, Yongrui Chen, and Guilin Qi. 2023. Evaluation of ChatGPT as a question answering system for answering complex questions. arXiv preprint arXiv:2303.07992 (2023). · arXiv
- Guido Zuccon and Bevan Koopman. 2023. Dr ChatGPT, tell me what I want to hear: How prompt knowledge impacts health answer correctness. arXiv preprint arXiv:2302.13793 (2023). · arXiv
Cited by
The power of noise: Redefining retrieval for rag systems
Cuconasu, F., Trappolini, G., Siciliano, F., Filice, S., Campagnano, C., Maarek, Y., Tonellotto, N., & Silvestri, F. (2024) · Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval · doi:10.1145/3626772.3657834 · Cited by 236
AI Chatbots in Education: Challenges and Opportunities
Davar, N. F., Dewan, M. A. A., & Zhang, X. (2025) · Information · doi:10.3390/info16030235 · Cited by 123
An automated framework for assessing how well LLMs cite relevant medical references
Wu, K., Wu, E., Wei, K., Zhang, A., Casasola, A., Nguyen, T., Riantawan, S., Shi, P., Ho, D., & Zou, J. (2025) · Nature Communications · doi:10.1038/s41467-025-58551-6 · Cited by 72
Zhao, Q., Wang, R., Cen, Y., Zha, D., Tan, S., Dong, Y., & Tang, J. (2024) · Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2024.emnlp-main.1259 · Cited by 32
Metadata
- Authors
- Zuccon, G., Koopman, B., & Shaik, R.
- Venue
- Proceedings of the Annual International ACM SIGIR Conference on Research and Development in Information Retrieval in the Asia Pacific Region
- Publisher
- ACM
- Published
- 2023-11-26
- Pages
- 46-51
- OpenAlex
- W4388955808
- Cited by
- 51
- References
- 12
- FWCI
- 9.64
- Type
- conference paper
- Language
- en
- Open access
- bronze
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: ChatGPT Hallucinates when Attributing Answers. https://app.cochat.ai/papers/10.1145/3624918.3625329
- The paper
- Zuccon, G., Koopman, B., & Shaik, R. (2023). ChatGPT Hallucinates when Attributing Answers. Proceedings of the Annual International ACM SIGIR Conference on Research and Development in Information Retrieval in the Asia Pacific Region. https://doi.org/10.1145/3624918.3625329
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.