Survey of Hallucination in Natural Language Generation
Open accessThe DOI matches its Crossref record; 355 of the 360 references it cites were found in a registry or an index, and 4 are not checked yet; 224 of the 256 sources its claims cite were checked for support.
Abstract
Natural Language Generation (NLG) has improved exponentially in recent years thanks to the development of sequence-to-sequence deep learning technologies such as Transformer-based language models. This advancement has led to more fluent and coherent NLG, leading to improved development in downstream tasks such as abstractive summarization, dialogue generation, and data-to-text generation. However, it is also apparent that deep learning based generation is prone to hallucinate unintended text, which degrades the system performance and fails to meet user expectations in many real-world scenarios. To address this issue, many studies have been presented in measuring and mitigating hallucinated texts, but these have never been reviewed in a comprehensive manner before. In this survey, we thus provide a broad overview of the research progress and challenges in the hallucination problem in NLG. The survey is organized into two parts: (1) a general overview of metrics, mitigation methods, and future directions, and (2) an overview of task-specific research progress on hallucinations in the following downstream tasks, namely abstractive summarization, dialogue generation, generative question answering, data-to-text generation, and machine translation. This survey serves to facilitate collaborative efforts among researchers in tackling the challenge of hallucinated texts in NLG.
Verified 2026-11-19 · 4 references not checked yet
Overview
CoChat Summary
TL;DR
This survey provides a broad overview of hallucination in natural language generation, covering definitions, contributors, metrics, mitigation methods, and task-specific progress across summarization, dialogue, GQA, data-to-text, machine translation, vision-language generation, and LLMs. It distinguishes intrinsic from extrinsic hallucination and argues that hallucination must be treated differently across tasks and sources.
Key findings
- The survey categorizes hallucination into intrinsic hallucination, where generated output contradicts the source, and extrinsic hallucination, where output cannot be verified from the source; extrinsic hallucination is not always erroneous because it may come from factually correct external information.
- It reports prevalence examples from cited work: 25% of summaries generated by state-of-the-art abstractive summarization systems have hallucinated content; 62% of first sentences in WIKIBIO have additional information not stated in the corresponding infobox; and 74% of samples in the QMSum dataset consist of inconsistent facts.
- It reviews hallucination metrics as statistical, model-based (IE-based, QA-based, NLI-based, faithfulness classification, LM-based), and human evaluation, and mitigation methods as data-related and modeling/inference methods including architecture changes, planning/sketching, RL, multi-task learning, controllable generation, and post-processing.
- For LLMs, it notes that RLHF reduced the hallucination rate from 41% to 21% in closed-domain tasks for InstructGPT, and discusses reference-dependent and reference-free hallucination metrics plus mitigation via data refinement, RAG, safety fine-tuning, model editing, decoding methods, chain-of-thought variants, post-processing, and ensembling.
Methods
- Literature survey design: the paper reviews existing research on hallucination in NLG rather than conducting new experiments or primary data collection.
- Coverage: general NLG hallucination topics (definitions, contributors, metrics, mitigation, future directions) plus task-specific sections on abstractive summarization, dialogue generation, generative question answering, data-to-text generation, neural machine translation, vision-language generation, and large language models.
- Analysis: synthesizes and compares definitions, metrics, mitigation approaches, and future directions across tasks, with examples in tables and figures; the LLM section was updated in Jan 2024.
Limitations
- The paper states that GQA has no standard definition or categorization of hallucination and no automatic metric specifically for hallucination in GQA; it also notes that open-domain dialogue lacks a standard hallucination metric and that VL hallucination research is still in its early stage.
- The paper states that extrinsic hallucination has been somewhat overlooked and is more challenging to reduce, and that most existing mitigation methods do not distinguish between intrinsic and extrinsic hallucination.
- Evident from the text: as a survey, it does not provide new empirical validation of metrics or mitigation methods, and its LLM coverage is explicitly updated only through Jan 2024, so later work is not included.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures
Verification
Claims the paper makes
124 Valid · 26 Not valid · 19 Not checked · 135 Own results and interpretations
At least one cited source doesn’t support this claim or points to a different paper, and none confirmed it — see the verdicts below.
In the paper
“Furthermore, perturbation, where the insertions of most common tokens are placed at the beginning of the sentence, seems to be the most successful in hallucination mitigation.”
What the sources say
- Not checked — no full text of the sourcecites [133] [133] Katherine Lee, Orhan Firat, Ashish Agarwal, Clara Fannjiang, and David Sussillo. 2019. Hallucinations in Neural Machine Translation. ICLR (2019). -
- Disputedcites [215] [215] Vikas Raunak, Arul Menezes, and Marcin Junczys-Dowmunt. 2021. The Curious Case of Hallucinations in Neural Machine Translation. (4 2021), 1172-1183. https://arxiv.org/abs/2104.06683v1 - sourceExperimental — accuracy not yet measured
In the passage describing the perturbation procedure (page unavailable), the paper says that inserting common tokens at the beginning is used to generate hallucinations, not mitigate them. Its prevention section instead discusses approaches such as robust learning and corpus-level filtering. (LLM, conf=0.92)
Checked by a reasoning model — may be wrong.
hedgedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
At least one cited source doesn’t support this claim or points to a different paper, and none confirmed it — see the verdicts below.
In the paper
“While there are works on the visual extension of faithfulness-oriented RLHF [303], these essentially involve labeling more data, which is expensive and difficult to scale.”
What the sources say
- Disputedcites [303] [303] Tianyu Yu, Yuan Yao, Haoye Zhang, Taiwen He, Yifeng Han, Ganqu Cui, Jinyi Hu, Zhiyuan Liu, Hai-Tao Zheng, Maosong Sun, and Tat-Seng Chua. 2023. RLHF-V: Towards Trustworthy MLLMs via Behavior Alignment from Fine-grained Correctional Human Feedback. CoRR abs/2312.00849 (2023). https://doi.org/10.48550/ARXIV.2312.00849 arXiv:2312.00849 - sourceExperimental — accuracy not yet measured
RLHF-V does use human annotation for visual-model alignment, but it makes the contrary point about needing more labeled data. The Fine-grained Correctional Human Preference Collection section says its corrections improve data efficiency, and Section 4.3 reports comparable hallucination rates using only 200 preference examples versus an order of magnitude more ranking-labeled examples. Our copy does not establish a separate claim about annotation cost or scalability. (LLM, conf=0.88)
Checked by a reasoning model — may be wrong.
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Disputed
Every source the paper cites for this claim was checked and supports it.
In the paper
“Nevertheless, as illustrated by Falke et al. [66], off-the-shelf NLI models tend to transfer poorly to the abstractive summarization task.”
What the sources say
- Supportedcites [66] [66] Tobias Falke, Leonardo FR 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 Linguistics . 2214-2220. - source
The Introduction states that models trained on NLI datasets transfer poorly to the paper’s downstream task of ranking generated summaries by correctness. The Conclusions likewise say that out-of-the-box NLI models do not perform well on that task. (LLM, conf=0.94)
- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Such wrong association results in generation with facts mixed up between two similar entities [60, 233].”
What the sources say
- Partly supportedcites [60] [60] Nouha Dziri, Andrea Madotto, Osmar Zaiane, and Avishek Joey Bose. 2021. Neural Path Hunter: Reducing Hallucination in Dialogue Systems via Path Grounding. EMNLP (2021). - sourceExperimental — accuracy not yet measured
The GPT2-KG passage describes an incorrect association between entities, and G Error Analysis gives an example of a generated response assigning a book to the wrong author. These support the claim that facts can be mixed up, but the paper does not say the entities involved are similar. The supplied passages do not provide page numbers. (LLM, conf=0.83)
- Supportedcites [233] [233] Kurt Shuster, Spencer Poff, Moya Chen, Douwe Kiela, and Jason Weston. 2021. Retrieval Augmentation Reduces Hallucination in Conversation. EMNLP (2021). - source
The Introduction states that language models often mix up facts between two similar entities when generating factually incorrect statements. This directly supports the cited description of the resulting error. (LLM, conf=0.92)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“As a result, the target reference may contain information that cannot be supported by the source [198, 268].”
What the sources say
- Supportedcites [198] [198] 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) . 1173-1186. - source
The Dataset section says that reference targets selected from web sentences may contain information that cannot be inferred from the source. The Data Annotation Process section also describes deleting phrases unsupported by table cells. (LLM, conf=0.93)
- Supportedcites [268] [268] Hongmin Wang. 2019. Revisiting Challenges in Data-to-Text Generation with Fact Grounding. In Proceedings of the 12th International Conference on Natural Language Generation . 311-322. - source
The Introduction reports that about 40% of human-written game-summary content cannot be directly mapped to the input boxscore records and draws on information beyond those inputs. Section 2.1 likewise describes reference content not licensed by the boxscore table. This supports the claim that a target reference may contain information the source does not support. (LLM, conf=0.93)
hedgedstates a possible cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
117 more claims · 74 with their cited sources checked
Create a free account to see all 304 claims and to verify the 2 references still open.
- Borderline: one of the tests for a substantial claim is unclear. 1
- It may state two separate claims as one. 90
- Probably not a substantial claim. 142
- The quote only partly states this claim. 1
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Hallucination can lead to serious consequences such as incorrect or misleading code, security risks, and ethical issues [186].”
From A survey on large language model based autonomous agents (2024) · cites this paper as [186]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“citation fabrication constitutes a form of extrinsic hallucination in the absence of retrieval augmentation, since the model generates references not grounded in any provided source material [4].”
From How LLMs Cite and Why It Matters: A Cross-Model Audit of Reference Fabrication in AI-Assisted Academic Writing and Methods to Detect Phantom Citations (2026) · cites this paper as [4]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Recent taxonomic work distinguishes at least seven cases [15,18].”
From Hallucination as epistemic pollution: epistemic injustice and credibility in futures-oriented science education (2026) · cites this paper as [18]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“LLMs struggle to generate accurate and up-to-date responses on current topics, due to outdated knowledge, lack of domain-specific expertise, or hallucination13–17.”
From Leveraging long context in retrieval augmented language models for medical question answering (2025) · cites this paper as [16]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Fine-tuning can enrich LLMs with domain knowledge, enhance their specific ability, improve the fairness and reliability of the outputs, and prevent certain damage caused by hallucination [26].”
From FederatedScope-LLM: A Comprehensive Package for Fine-tuning Large Language Models in Federated Learning (2024) · cites this paper as [26]
Experimental — accuracy not yet measured
41 more claims about this paper
Create a free account to see every claim other papers make about this one.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Extrinsic Object Hallucination
“CHAIR (Caption Hallucination Assessment with Image Relevance) metric, which calculates what proportion of object words generated are actually in the image according to the ground truth captions.”
§ Extrinsic Object Hallucination
“CHAIR 𝑖 measures per-instance object hallucination, i.e. what fraction of object instances in each generated caption are hallucinated.”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ 7.1 Hallucination Definition in Abstractive Summarization
“For instance, in Table 1, given the input article shown in the caption, an example of intrinsic hallucination is ' The Ebola vaccine was rejected by the FDA in 2019, '”
§ 2.1 Categorization
“in the abstractive summarization task from Table 1, the generated summary ' The first Ebola vaccine was approved in 2021 ' contradicts the source content”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Survey of Hallucination in Natural Language Generation
“The survey is organized into two parts: (1) a general overview of metrics, mitigation methods, and future directions; (2) an overview of task-specific research progress on hallucinations in the following downstream tasks, namely abstractive summarization, dialogue generation, generative question answering, data-to-text generation, machine translation, and visual-language generation; and (3) hallucinations in large language models (LLMs)”
§ 1 INTRODUCTION
“The third part discusses this phenomenon in LLMs in Section 13.”
Citations
Cites
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · 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) · doi:10.18653/v1/n19-1423 · Cited by 54,325 · cited 3×
“Recently, the rapid development of NLG has captured the imagination of many thanks to the advances in deep learning technologies, especially Transformer [189]-based models like BERT [29], BART [100], GPT-2 [149], and GPT-3 [16].”
BackgroundAn Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., et al. (2020) · arXiv · doi:10.48550/arxiv.2010.11929 · Cited by 47,472
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
Sequence to Sequence Learning with Neural Networks
Sutskever, I., Vinyals, O., & Le, Q. V. (2014) · arXiv · doi:10.48550/arxiv.1409.3215 · Cited by 17,766
Advances in Neural Information Processing Systems 29
Teymur, O., Zygalakis, K., & Calderhead, B. (2016) · Cited by 13,887
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., Zhou, Y., Li, W., & Liu, P. J. (2019) · arXiv · doi:10.48550/arxiv.1910.10683 · Cited by 10,929 · cited 1×
“This work investigates the performances of GPT-3 [13], GPT-Neo/J [141], GPT-2 [105] and a T5-based model [106].”
BackgroundAdvances in Neural Information Processing Systems 14: Proceedings of the 2001 Conference
(2002) · The MIT Press eBooks · doi:10.7551/mitpress/1120.001.0001 · Cited by 8,890
Lewis, M., Liu, Y., Goyal, N., Ghazvininejad, M., Mohamed, A., Levy, O., Stoyanov, V., & Zettlemoyer, L. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.703 · Cited by 8,821 · cited 1×
“Recently, the development of NLG has conquered the imagination of many thanks to the advances in deep learning technologies, especially the Transformer [159]-based models like BERT [24], BART [83], GPT-2 [127], and GPT3 [13].”
BackgroundLewis, M., Liu, Y., Goyal, N., Ghazvininejad, M., Mohamed, A., Levy, O., Stoyanov, V., & Zettlemoyer, L. (2019) · arXiv · doi:10.48550/arxiv.1910.13461 · Cited by 8,816 · cited 1×
“Recently, the development of NLG has conquered the imagination of many thanks to the advances in deep learning technologies, especially the Transformer [159]-based models like BERT [24], BART [83], GPT-2 [127], and GPT3 [13].”
Background
Research works printed but not linked: 16
- [4] Joshua Albrecht and Rebecca Hwa. 2007. A Re-examination of Machine Learning Approaches for Sentence-Level MT Evaluation. In Proceedings of the 45th Annual Meeting of the Association of Computational Linguistics . Association for Computational Linguistics, Prague, Czech Republic, 880-887. https://aclanthology.org/P07-1111 -
- [16] Andrzej Białecki, Robert Muir, Grant Ingersoll, and Lucid Imagination. 2012. Apache lucene 4. In SIGIR 2012 workshop on open source information retrieval . 17. -
- [53] Yilun Du, Shuang Li, Antonio Torralba, Joshua B Tenenbaum, and Igor Mordatch. 2023. Improving Factuality and Reasoning in Language Models through Multiagent Debate. In Forty-first International Conference on Machine Learning . -
- [121] Philipp Koehn and Rebecca Knowles. 2017. Six Challenges for Neural Machine Translation. In First Workshop on Neural Machine Translation . Association for Computational Linguistics, 28-39. -
- [145] Junnan Li, Dongxu Li, Caiming Xiong, and Steven Hoi. 2022. BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation. In ICML . -
- [149] Tian Li, Ahmad Beirami, Maziar Sanjabi, and Virginia Smith. 2020. Tilted Empirical Risk Minimization. In International Conference on Learning Representations . -
- [164] Jiasen Lu, Jianwei Yang, Dhruv Batra, and Devi Parikh. 2018. Neural Baby Talk. In CVPR . -
- [165] Yukun Ma, Khanh Linh Nguyen, Frank Z Xing, and Erik Cambria. 2020. A Survey on Empathetic Dialogue Systems. Information Fusion 64 (2020), 50-70. -
- [175] Marianna J. Martindale, Marine Carpuat, Kevin Duh, and Paul McNamee. 2019. Identifying Fluently Inadequate Output in Neural and Statistical Machine Translation. In MTSummit . -
- [179] Kathleen McKeown. 1992. Text Generation . Cambridge University Press. -
- [209] Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, and Ilya Sutskever. 2019. Language Models are Unsupervised Multitask Learners. OpenAI Blog 1, 8 (2019), 9. -
- [238] Hui Su, Xiaoyu Shen, Sanqiang Zhao, Zhou Xiao, Pengwei Hu, Cheng Niu, and Jie Zhou. 2020. Diversifying Dialogue Generation with Non-Conversational Text. In 58th Annual Meeting of the Association for Computational Linguistics . ACL, 7087-7097. -
- [256] Zhaopeng Tu, Yang Liu, Lifeng Shang, Xiaohua Liu, and Hang Li. 2017. Neural Machine Translation with Reconstruction. In Thirty-First AAAI Conference on Artificial Intelligence . -
- [261] Neeraj Varshney, Wenlin Yao, Hongming Zhang, Jianshu Chen, and Dong Yu. 2023. A Stitch in Time Saves Nine: Detecting and Mitigating Hallucinations of LLMs by Validating Low-Confidence Generation. CoRR abs/2307.03987 (2023). https://doi.org/10.48550/arXiv.2307.03987 arXiv:2307.03987 - · arXiv · doi:10.48550/arxiv.2307.03987 · Paper page
- [285] Zeqiu Wu, Michel Galley, Chris Brockett, Yizhe Zhang, Xiang Gao, Chris Quirk, Rik Koncel-Kedziorski, Jianfeng Gao, Hannaneh Hajishirzi, Mari Ostendorf, et al. 2021. A Controllable Model of Grounded Response Generation. In Proceedings of the AAAI Conference on Artificial Intelligence , Vol. 35. 14085-14093. -
- [291] Xinnuo Xu, Ondřej Dušek, Verena Rieser, and Ioannis Konstas. 2021. AGGGEN: Ordering and Aggregating while Generating. roceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL2021) (2021). -
Cited by
Items 601–700 of 4,844
Jill Watson: Scaling and Deploying an AI Conversational Agent in Online Classrooms
Kakar, S., Maiti, P., Taneja, K., Nandula, A., Nguyen, G., Zhao, A., Nandan, V., & Goel, A. K. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-63028-6_7 · Cited by 48
Wang, H., Zhao, S., Qiang, Z., Li, Z., Liu, C., Xi, N., Du, Y., Qin, B., & Liu, T. (2025) · ACM Transactions on Knowledge Discovery from Data · doi:10.1145/3686807 · Cited by 65
“Their general domain knowledge often falls short when addressing such specialized fields, where accurate and domain-specific expert knowledge is critical, which leads to hallucination (Ji et al. 2023) in the model responses, especially for languages that are less well-trained than English.”
Chern, I.-C., Chern, S., Chen, S., Yuan, W., Feng, K., Zhou, C., He, J., Neubig, G., & Liu, P. (2023) · arXiv · doi:10.48550/arxiv.2307.13528 · Cited by 29
Influence of External Information on Large Language Models Mirrors Social Cognitive Patterns
Bian, N., Lin, H., Liu, P., Lu, Y., Zhang, C., He, B., Han, X., & Sun, L. (2023) · arXiv · doi:10.48550/arxiv.2305.04812 · Cited by 24
Safrai, M., & Orwig, K. E. (2024) · Journal of Assisted Reproduction and Genetics · doi:10.1007/s10815-024-03089-7 · Cited by 35
“In the world of AI, those are defined as hallucina-tions—a confident response by artificial intelligence that does not seem to be justified by its training data [28].”
BackgroundChen, L., Jiang, Z., Xia, D., Cai, Z., Sun, L., Childs, P., & Zuo, H. (2024) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613904.3642887 · Cited by 42
Low Resource Summarization using Pre-trained Language Models
Munaf, M., Afzal, H., Mahmood, K., & Iltaf, N. (2024) · ACM Transactions on Asian and Low-Resource Language Information Processing · doi:10.1145/3675780 · Cited by 40
“Despite its popularity and intelligent generative ability, it suffers from hallucinations [19] and factual inaccuracy [5], is sensitive to input phrases, 6 fails on basic math tasks [5], and is yet to perform complex code analyses.”
A conversational agent for creating automations exploiting large language models
Gallo, S., Paternò, F., & Malizia, A. (2024) · Personal and Ubiquitous Computing · doi:10.1007/s00779-024-01825-5 · Cited by 29
“For example, Ji et al. [19] distinguish between “intrinsic” (the generated output contradicts the source content) and “extrinsic” (the generated output cannot be verified from the source content) hallucinations, while Zhang et al. [20] propose a classification into “input-conflicting” (LLMs…”
Preventing and Detecting Misinformation Generated by Large Language Models
Liu, A., Sheng, Q., & Hu, X. (2024) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3626772.3661377 · Cited by 68
Analyzing digital communication: a comprehensive literature review
Strauss, C., Harr, M. D., & Pieper, T. M. (2024) · Management Review Quarterly · doi:10.1007/s11301-024-00455-8 · Cited by 76
“In addition, it is important to note that AI systems can hallucinate , meaning they may produce incorrect results based solely on probabilities (e.g., Hagendorff 2023; Ji et al. 2023; Raji et al. 2022).”
Behind the Scenes: A Critical Perspective on GenAI and Open Educational Practices
Aksoy, D. A., & Kurşun, E. (2024) · Open Praxis · doi:10.55982/openpraxis.16.3.674 · Cited by 31
Wang, S., Cooper, N., & Eby, M. R. (2024) · Big Data & Society · doi:10.1177/20539517241290220 · Cited by 43
To What Extent Have LLMs Reshaped the Legal Domain So Far? A Scoping Literature Review
Padiu, B., Iacob, R. C. A., Rebedea, T. E., & Dascălu, M. (2024) · Information · doi:10.3390/info15110662 · Cited by 58
Lee, J., Cha, H., Hwangbo, Y., & Cheon, W. (2024) · Journal of Personalized Medicine · doi:10.3390/jpm14121131 · Cited by 31
Matsumoto, N., Choi, H., Moran, J., Hernandez, M. E., Venkatesan, M., Li, X., Chang, J.-H., Wang, P., & Moore, J. H. (2025) · Bioinformatics · doi:10.1093/bioinformatics/btaf031 · Cited by 23
Visual knowledge in the big model era: retrospect and prospect
Wang, W., Yang, Y., & Pan, Y. (2025) · Frontiers of Information Technology & Electronic Engineering · doi:10.1631/fitee.2400250 · Cited by 47
“For instance, big models may generate plausible-sounding but factually incorrect or nonsensical answers, known as “hallucination”, reflecting a surface-level mimicry of human output rather than genuine comprehension (Ji et al, 2023).”
HaVen: Hallucination-Mitigated LLM for Verilog Code Generation Aligned with HDL Engineers
Yang, Y., Teng, F., Liu, P., Qi, M., Lv, C., Li, J., Zhang, X., & He, Z. (2025) · Design, Automation, and Test in Europe · doi:10.23919/date64628.2025.10993072 · Cited by 59
“Hallucination [11] in language models refers to the generation of incorrect or misleading information not grounded in reality.”
Wei, L., Yu, Y., Qin, Y., & Zhang, S. (2025) · Information · doi:10.3390/info16080656 · Cited by 49
The trust crisis in artificial intelligence: AI hallucinations and human-AI collaboration
Cheng, Q., Dai, Y., Liu, X., & Peng, S. (2026) · Technology in Society · doi:10.1016/j.techsoc.2026.103286 · Cited by 34
A study on the performance of ChatGPT in infectious diseases clinical consultation
Sarink, M. J., Bakker, I. L., Anas, A. A., & Yusuf, E. (2023) · Clinical Microbiology and Infection · doi:10.1016/j.cmi.2023.05.017 · Cited by 24
The Use of Artificial Intelligence Tools to Prepare Medical School Applications
Hashimoto, D. A., & Johnson, K. B. (2023) · Academic Medicine · doi:10.1097/acm.0000000000005309 · Cited by 23
Navigating Generative AI: The Teacher Librarian's Role in Cultivating Ethical and Critical Practices
Oddone, K., Garrison, K., & Gagen-Spriggs, K. (2023) · Journal of the Australian Library and Information Association · doi:10.1080/24750158.2023.2289093 · Cited by 38
“This is particularly important given ChatGPT ’ s tendency to ‘ hallucinate ’ or invent a convincing yet inaccurate response to a prompt (Ji et al., 2023).”
BackgroundGeneration Probabilities Are Not Enough: Uncertainty Highlighting in AI Code Completions
Vasconcelos, H., Bansal, G., Fourney, A., Liao, Q. V., & Vaughan, J. W. (2024) · ACM Transactions on Computer-Human Interaction · doi:10.1145/3702320 · Cited by 91
(2025) · doi:10.1007/s12525-025-00862-z · Cited by 23
“…operators’ ability to govern the evolution of models, fine-tune them for specific domains (e.g., multi-sided platforms, digital matching services), and mitigate propagated bias in marketplace decisions (e.g., pricing, matching, recommendation) (Fetzer et al., 2025; Ji et al., 2023; Ray, 2023).”
Generation Probabilities Are Not Enough: Uncertainty Highlighting in AI Code Completions
Vasconcelos, H., Bansal, G., Fourney, A., Liao, Q. V., & Vaughan, J. W. (2023) · arXiv · doi:10.48550/arxiv.2302.07248 · Cited by 23
Mindful-RAG: A Study of Points of Failure in Retrieval Augmented Generation
Agrawal, G., Kumarage, T., Alghamdi, Z., & Liu, H. (2024) · arXiv · doi:10.48550/arxiv.2407.12216 · Cited by 23
He, G., Demartini, G., & Gadiraju, U. (2025) · Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3706598.3713218 · Cited by 23
Memolet: Reifying the Reuse of User-AI Conversational Memories
Yen, R., & Zhao, J. (2024) · Proceedings of the 37th Annual ACM Symposium on User Interface Software and Technology · doi:10.1145/3654777.3676388 · Cited by 22
Sufi, F. K. (2024) · Sustainability · doi:10.3390/su16072742 · Cited by 29
Dierickx, L., Sirén‐Heikel, S., & Lindén, C. (2024) · Emerging Media · doi:10.1177/27523543241288846 · Cited by 38
“Arti fi cial hallucinations, which relate to the generation of unintended fabricated content, challenge the credibility and reliability of generated content in fact-checking contexts (Cuartielles et al., 2023; Dwivedi et al., 2023; Ji et al., 2023; Jones et al., 2023; Rawte et al., 2023).”
Liao, D., Pan, S., Sun, X., Ren, X., Huang, Q., Xing, Z., Jin, H., & Li, Q. (2024) · IEEE Transactions on Software Engineering · doi:10.1109/tse.2024.3486195 · Cited by 23
Hallucinations of large multimodal models: Problem and countermeasures
Sun, S., Lin, Z., & Wu, X. (2025) · Information Fusion · doi:10.1016/j.inffus.2025.102970 · Cited by 38
Scaffolding middle school mathematics curricula with large language models
Malik, R., Abdi, D., Wang, R., & Demszky, D. (2025) · British Journal of Educational Technology · doi:10.1111/bjet.13571 · Cited by 44
“Researchers have noted that these models can sometimes generate unreliable solutions to math problems (Frieder et al., 2024) and may “hallucinate” information, producing content that appears plausible but is not actually accurate (Ji et al., 2023).”
Li, H., Huang, J., Ji, M., Yang, Y., & An, R. (2025) · Journal of Medical Internet Research · doi:10.2196/66098 · Cited by 46
“However, they also tend to generate plausible but incorrect or nonsensical information not grounded in the input data or real-world knowledge, commonly referred to as hallucinations [8,9].”
ChatGPT is incredible (at being average)
Rudko, I., & Bonab, A. B. (2025) · Ethics and Information Technology · doi:10.1007/s10676-025-09845-2 · Cited by 32
“Because the problem of hallucinations is so evidently out there, researchers actively seek measures to ensure LLMs’ trustworthiness through sophisticated detection and prevention methods (Ji et al., 2023; Ye et al., 2023).”
Progress of AI-Driven Drug–Target Interaction Prediction and Lead Optimization
Wang, Q., Sun, B., Yi, Y., Velkov, T., Shen, J., Dai, C., & Jiang, H. (2025) · International Journal of Molecular Sciences · doi:10.3390/ijms262010037 · Cited by 38
“Therefore, future efforts should emphasize the systematic curation of both positive and negative results, along with the establishment of shared, standardized data infrastructures to enhance cross-study comparability and robustness [229].”
Yamazaki, T., Yoshikawa, K., Kawamoto, T., Mizumoto, T., Ohagi, M., & Sato, T. (2023) · Advanced Robotics · doi:10.1080/01691864.2023.2244554 · Cited by 33
“This type of inappropriate generation is a well-known challenge in language models, commonly referred to as Hallucination [20,21].”
BackgroundChatGPT: Opportunities, risks and priorities for psychiatry
Wei, Y., Guo, L., Lian, C., & Chen, J. (2023) · Asian Journal of Psychiatry · doi:10.1016/j.ajp.2023.103808 · Cited by 28
Automatic recognition of second language speech-in-noise
Kim, S.-E., Chernyak, B. R., Seleznova, O., Keshet, J., Goldrick, M., & Bradlow, A. R. (2024) · JASA Express Letters · doi:10.1121/10.0024877 · Cited by 38
Li, J., He, C., Hu, J., Jia, B., Halevy, A. Y., & Ma, X. (2024) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613904.3642853 · Cited by 31
FAIR Enough: Develop and Assess a FAIR-Compliant Dataset for Large Language Model Training?
Raza, S., Ghuge, S., Ding, C., Dolatabadi, E., & Pandya, D. A. (2024) · Data Intelligence · doi:10.1162/dint_a_00255 · Cited by 49
“Accuracy and Reliability : LLMs sometimes producing incorrect information highlights the need for effective fact-checking mechanisms [29].”
BackgroundDoLa: Decoding by Contrasting Layers Improves Factuality in Large Language Models
Chuang, Y.-S., Xie, Y., Luo, H., Kim, Y., Glass, J., & He, P. (2023) · arXiv · doi:10.48550/arxiv.2309.03883 · Cited by 29
“Hallucinations in LLMs. Hallucinations in LLMs refer to generated content not based on training data or facts, caused by various factors like imperfect learning and decoding (Ji et al., 2023).”
BackgroundMethodVeriPlan: Integrating Formal Verification and LLMs into End-User Planning
Lee, C. P., Porfirio, D., Wang, X. J., Zhao, K. C., & Mutlu, B. (2025) · Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3706598.3714113 · Cited by 21
Patient-Friendly Discharge Summaries in Korea Based on ChatGPT: Software Development and Validation
Kim, H., Jin, H. M., Jung, Y. B., & You, S. C. (2024) · Journal of Korean Medical Science · doi:10.3346/jkms.2024.39.e148 · Cited by 32
“22,23 Since there is evidence that automatic metrics (such as ROUGE, 24 BERTScore, 25 METEOR, 26 and BLEU 27 ) cannot effectively evaluate the factuality of text summarization, 17,22 we conducted a human evaluation of the factuality score to assess the level of hallucination in our study.”
Uranbey, Ö., Özbey, F., Kaygisiz, Ö. F., & Ayrancı, F. (2024) · Cureus · doi:10.7759/cureus.58607 · Cited by 33
“Hallucination is one particularly noteworthy concern [21].”
Fok, R., Lipka, N., Sun, T., & Siu, A. F. (2024) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613904.3641969 · Cited by 37
Effectiveness of Generative AI for Post-Earthquake Damage Assessment
Estêvão, J. M. C. (2024) · Buildings · doi:10.3390/buildings14103255 · Cited by 33
Choi, H., Lee, D., Kang, Y.-K., & Suh, M. (2025) · European Journal of Nuclear Medicine and Molecular Imaging · doi:10.1007/s00259-025-07101-9 · Cited by 25
Agentic Workflows for Improving Large Language Model Reasoning in Robotic Object-Centered Planning
Moncada-Ramirez, J., Matez-Bandera, J.-L., González-Jiménez, J., & Ruiz-Sarmiento, J.-R. (2025) · Robotics · doi:10.3390/robotics14030024 · Cited by 35
“However, a major disadvantage of relying on LLMs is their tendency to produce incorrect, inaccurate, or invented information, a phenomenon known as hallucination [20] that can potentially lead to erroneous robot behavior.”
Few-shot learning of geological cross-sections from sparse data using large language model
Li, H., & Shi, C. (2025) · Geodata and AI. · doi:10.1016/j.geoai.2025.100010 · Cited by 27
Asking generative artificial intelligence the right questions improves writing performance
Cheng, Y., Fan, Y., Li, X., Chen, G., Gašević, D., & Swiecki, Z. (2025) · Computers and Education: Artificial Intelligence · doi:10.1016/j.caeai.2025.100374 · Cited by 21
A Systematic Literature Review of Hallucinations in Large Language Models
Woesle, C., Fischer-Brandies, L., & Buettner, R. (2025) · IEEE Access · doi:10.1109/access.2025.3601206 · Cited by 40
“Moreover, Ji et al. [9] emphasize that mitigation techniques vary greatly in effectiveness across domains and models, and that there is limited consensus in the literature regarding the most effective or generalizable solutions.”
InfluentialChen, F., Li, Y., Chen, Y., Bian, Z., Duo, L., Zhou, Q., Zhang, L., & ADVANCED Working Group (2025) · Journal of Evidence-Based Medicine · doi:10.1111/jebm.70075 · Cited by 34
“Traditional metrics are insufficient for comprehensively quantifying the levels of model hallucination [74].”
A comprehensive survey on automatic text summarization with exploration of LLM-based methods
Zhang, Y., Jin, H., Meng, D., Wang, J., & Tan, J. (2025) · Neurocomputing · doi:10.1016/j.neucom.2025.131928 · Cited by 324
“Generating summaries directly via LLMs can encounter hal-lucination issues, resulting in factually incorrect information[169].”
Karakurt, E., & Akbulut, A. (2025) · Applied Sciences · doi:10.3390/app16010368 · Cited by 66
Ilicki, J. (2023) · Mayo Clinic Proceedings: Digital Health · doi:10.1016/j.mcpdig.2023.03.006 · Cited by 20
“1,4,15 Manufacturers must ensure that a medical LLM software performs in a safe and predictable manner according to relevant legislation (eg, the Medical Device Regulation in Europe).”
Naikar, N., Brady, A., Moy, G., & Kwok, H.-W. (2023) · Ergonomics · doi:10.1080/00140139.2023.2281898 · Cited by 35
Enhancing Uncertainty-Based Hallucination Detection with Stronger Focus
Zhang, T., Qiu, L., Guo, Q., Deng, C., Zhang, Y., Zhang, Z., Zhou, C., Wang, X., & Fu, L. (2023) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2023.emnlp-main.58 · Cited by 160
From text to image: challenges in integrating vision into ChatGPT for medical image interpretation
Koga, S., & Du, W. (2024) · Neural Regeneration Research · doi:10.4103/nrr.nrr-d-24-00165 · Cited by 21
“Additionally, several studies pointed out difficulties in predicting accurate diagnoses from clinical information, including instances of inconsistency and inaccuracy (Ji et al., 2023; Koga, 2023; Thirunavukarasu et al., 2023).”
BackgroundA reliable knowledge processing framework for combustion science using foundation models
Sharma, V., & Raman, V. (2024) · Energy and AI · doi:10.1016/j.egyai.2024.100365 · Cited by 31
“This is a known problem of hallucination [23] caused due to knowledge gap and model’s ability to generate seemingly credible text based on patterns it has acquired from its training data, regardless of whether the generated content conforms to actual facts.”
BackgroundNatural Language Dataset Generation Framework for Visualizations Powered by Large Language Models
Ko, H.-K., Jeon, H., Park, G., Kim, D. H., Kim, N. W., Kim, J., & Seo, J. (2024) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613904.3642943 · Cited by 63
“This step is crucial, as large language models (LLMs) have been known to produce hallucinations in response to mathematical problems [27].”
Artificial intelligence, human cognition, and conscious supremacy
Mogi, K. (2024) · Frontiers in Psychology · doi:10.3389/fpsyg.2024.1364714 · Cited by 43
“One of the problems with LLMs, such as ChatGPT, is the occurrence of hallucination (Ji et al., 2023) and the tendency to produce sentences inconsistent with accepted facts, a term criticized by some researchers as an instance of anthropomorphism.”
BackgroundWould you trust an AI chef? Examining what people think when AI becomes creative with food
Califano, G., Zhang, T., & Spence, C. J. (2024) · International Journal of Gastronomy and Food Science · doi:10.1016/j.ijgfs.2024.100973 · Cited by 24
Generative AI: Overview, Economic Impact, and Applications in Asset ManagementPreprint
Luk, M. (2023) · SSRN Electronic Journal · doi:10.2139/ssrn.4574814 · Cited by 26
Synthetic Socio-Technical Systems: Poiêsis as Meaning Making
Bisconti, P., McIntyre, A., & Russo, F. (2024) · Philosophy & Technology · doi:10.1007/s13347-024-00778-0 · Cited by 21
“As evidenced by research in computer science and AI, DNNs often lack the precision and stability desired (del Campo & Leach, 2022 ; Ji et al., 2023 ).”
Explaining the Wait: How Justifying Chatbot Response Delays Impact User Trust
Zhang, Z., Tsiakas, K., & Schneegass, C. (2024) · ACM Conference on Conversational User Interfaces (CUI) · doi:10.1145/3640794.3665550 · Cited by 35
Antonie, N. I., Gheorghe, G., Ionescu, V. A., Tiucă, L.-C., & Diaconu, C. C. (2025) · Antibiotics · doi:10.3390/antibiotics14010060 · Cited by 29
“Such outputs can undermine trust in the system and pose challenges, particularly in high-stakes applications like medicine or legal analysis [55–57].”
Li, J., Ding, C., Liu, D., Chen, L., & Jiang, J. (2025) · Digital Discovery · doi:10.1039/d5dd00072f · Cited by 36
(2025) · Nature Reviews Bioengineering · doi:10.1038/s44222-025-00323-4 · Cited by 28
Ophthalmological Question Answering and Reasoning Using OpenAI o1 vs Other Large Language Models
Srinivasan, S., Ai, X., Zou, M., Zou, K., Kim, H., Lo, T. W. S., Pushpanathan, K., Yang, G. D., Goh, J. H. L., Kong, Y., et al. (2025) · JAMA Ophthalmology · doi:10.1001/jamaophthalmol.2025.2413 · Cited by 26
Tilton, A. K., Caplan, B. E., & Cole, B. J. (2025) · Frontiers in Digital Health · doi:10.3389/fdgth.2025.1616488 · Cited by 31
Sommer, F., Eberhard, B., & Holzinger, A. (2025) · Electronics · doi:10.3390/electronics14183583 · Cited by 31
“Recent state-of-the-art models have reduced, but not eliminated, this issue, prompting ongoing research into mitigation strategies including retrieval-augmented generation (RAG), verifier models, fact-checking pipelines, and uncertainty quantification [44].”
Chung, P., Swaminathan, A., Goodell, A. J., Kim, Y., Reincke, S. M., Han, L., Deverett, B., Sadeghi, M. A., Ariss, A.-B., Ghanem, M., et al. (2025) · NEJM AI · doi:10.1056/aidbp2500418 · Cited by 80
Benchmarking large language model-based agent systems for clinical decision tasks
Liu, Y., Carrero, Z. I., Jiang, X., Ferber, D., Wölflein, G., Zhang, L., Jayabalan, S., Lenz, T., Hui, Z., & Kather, J. N. (2026) · npj Digital Medicine · doi:10.1038/s41746-026-02443-6 · Cited by 37
“Previous research has suggested that hallucination is an inherent characteristic of languagegenerationbased onLLMs 28 , andagentsystemsmay amplifythese effects through repeated LLM calls.”
The Importance of Distrust in AI
Peters, T. M., & Visser, R. W. (2023) · Communications in computer and information science · doi:10.1007/978-3-031-44070-0_15 · Cited by 44
Lin, H., Luo, Z., Ma, J., & Chen, L. (2023) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/v1/2023.findings-emnlp.611 · Cited by 77
Jain, R. K., & Jain, A. (2024) · Lecture notes in networks and systems · doi:10.1007/978-3-031-66329-1_42 · Cited by 55
Making sense of negotiation and AI: The blossoming of a new collaboration
Falcão, H. (2024) · International Journal of Commerce and Contracting · doi:10.1177/20555636241269270 · Cited by 32
Arakawa, R., Lehman, J. F., & Goel, M. (2024) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3699759 · Cited by 45
Specification overfitting in artificial intelligence
Roth, B., de Araujo, P. H. L., Xia, Y., Kaltenbrunner, S., & Korab, C. (2024) · Artificial Intelligence Review · doi:10.1007/s10462-024-11040-6 · Cited by 26
“Limitations of these systems include the use of non-generalizable heuristics (Tu et al, 2020), generating texts that are biased, hateful, and toxic (Schick et al, 2021), and texts that look fluent and plausible but contain falsehoods and misinformation (Ji et al, 2023; Lin et al, 2022).”
Mindful-RAG: A Study of Points of Failure in Retrieval Augmented Generation
Agrawal, G., Kumarage, T., Alghamdi, Z., & Liu, H. (2024) · International Conference on Foundation and Large Language Models (FLLM) · doi:10.1109/fllm63129.2024.10852457 · Cited by 63
“In such cases, LLMs require additional relevant contextual knowledge through prompting.”
Equipping Llama with Google Query API for Improved Accuracy and Reduced HallucinationPreprint
Bae, Y. S., Kim, H. R., & Kim, J. H. (2024) · Research Square · doi:10.21203/rs.3.rs-4014474/v1 · Cited by 61
Opportunities and Risks of LLMs for Scalable Deliberation with Polis
Small, C. T., Vendrov, I., Durmus, E., Homaei, H., Barry, E., Cornebise, J., Suzman, T., Ganguli, D., & Megill, C. (2023) · arXiv · doi:10.48550/arxiv.2306.11932 · Cited by 20
“…et al., 2020; Kurita et al., 2019; Basta et al., 2019; Abid et al., 2021b; Sap et al., 2020), prompt injection attacks (Greshake et al., 2023; Perez and Ribeiro, 2022), so called “hallucinations” (LLMs making things up when they don’t know something; Ji et al., 2023b), and basic transparency.”
BackgroundVolcano: Mitigating Multimodal Hallucination through Self-Feedback Guided Revision
Lee, S., Park, S. H., Jo, Y., & Seo, M. (2024) · Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers) · doi:10.18653/v1/2024.naacl-long.23 · Cited by 20
HiTA: A RAG-Based Educational Platform that Centers Educators in the Instructional Loop
Liu, C., Hoang, L., Stolman, A., & Wu, B. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-64299-9_37 · Cited by 37
Multimodal Artificial Intelligence in Medicine
Judge, C. S., Krewer, F., O'Donnell, M. J., Kiely, L., Sexton, D. J., Taylor, G. W., Skorburg, J. A., & Tripp, B. (2024) · Kidney360 · doi:10.34067/kid.0000000000000556 · Cited by 28
Beyond Chatbot-K: On Large Language Models, “Generative AI,” and Rise of Chatbots—An Introduction
Goodlad, L. M. E., & Stone, M. (2024) · Critical AI · doi:10.1215/2834703x-11205147 · Cited by 36
Integrating AI Tutors in a Programming Course
Ma, I., Krone-Martins, A., & Lopes, C. V. (2024) · ACM Virtual Global Computing Education Conference (SIGCSE Virtual) · doi:10.1145/3649165.3690094 · Cited by 55
Classification of human- and AI-generated texts for different languages and domains
Schaaff, K., Schlippe, T., & Mindner, L. (2024) · International Journal of Speech Technology · doi:10.1007/s10772-024-10143-3 · Cited by 41
“Furthermore, texts generated by chat-bots often lack reliability which is often also referred to as hallucinations (Ji et al., 2023).”
Take Loads Off Your Developers: Automated User Story Generation using Large Language Model
Rahman, T., Zhu, Y., Maha, L., Roy, C. K., Roy, B., & Schneider, K. A. (2024) · IEEE International Conference on Software Maintenance (ICSM) · doi:10.1109/icsme58944.2024.00082 · Cited by 30
Chen, C., Li, S., So, A. D., Xu, Y., Guo, Z., Wang, X., Graham, D. W., & Zhu, Y. (2025) · Environmental Science & Technology · doi:10.1021/acs.est.4c07842 · Cited by 27
Self-citationThe ethical security of large language models: A systematic review
Liu, F., Jiang, J., Lu, Y., Huang, Z., & Jiang, J. (2025) · Frontiers of Engineering Management · doi:10.1007/s42524-025-4082-6 · Cited by 29
“These issues include hallucination of outputs (Cascella et al., 2023; Ji et al., 2023), potential misinformation and bias (Meyer et al., 2023; Zhou et al., 2023), risks of data privacy breaches (Su, 2024), and impacts on human autonomy (Ellis et al., 2024).”
Ranisch, R., & Haltaufderheide, J. (2025) · Digital Society · doi:10.1007/s44206-025-00161-2 · Cited by 21
Generative Artificial Intelligence in Medicine
Shanmugam, D., Agrawal, M., Movva, R., Chen, I. Y., Ghassemi, M., Jacobs, M. L., & Pierson, E. (2025) · Annual Review of Biomedical Data Science · doi:10.1146/annurev-biodatasci-103123-095332 · Cited by 42
Banerjee, A., Satish, A., & Wörndl, W. (2025) · Communications in computer and information science · doi:10.1007/978-3-031-87654-7_3 · Cited by 42
“Moreover, LLMs are prone to hallucinations, where they provide answers that contradict real-world knowledge or are irrelevant to user prompts [27, 64].”
Reliability engineering, risk management, and trustworthiness assurance for AI systems
Zhang, X., Wang, T., Ma, L., & Mahadevan, S. (2025) · Journal of Reliability Science and Engineering · doi:10.1088/3050-2454/adcef2 · Cited by 34
Dango: A Mixed-Initiative Data Wrangling System using Large Language Model
Chen, W., Tong, W., Case, A., & Zhang, T. (2025) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3706598.3714135 · Cited by 41
A survey on biomedical automatic text summarization with large language models
Huang, Z., Chen, X., Wang, Y., Huang, J., & Zhao, X. (2025) · Information Processing & Management · doi:10.1016/j.ipm.2025.104216 · Cited by 27
LLM-QL: A LLM-Enhanced Q-Learning Approach for Scheduling Multiple Parallel Drones
Zhou, Q. X., Wu, J., Zhu, M., Zhou, Y., Xiao, F., & Zhang, Y. (2025) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2025.3579386 · Cited by 33
“In LLMs, “hallucination” refers to the generation of over-estimated or underestimated values [45].”
Kassar, M., & Jizi, M. I. (2025) · Journal of Applied Accounting Research · doi:10.1108/jaar-05-2024-0175 · Cited by 58
Metadata
- Authors
- Ji, Z., Lee, N., Frieske, R., Yu, T., Su, D., Xu, Y., Ishii, E., Bang, Y. J., Madotto, A., & Fung, P.
- Affiliations
- Hong Kong University of Science and Technology (HK)
- ORCID
- Venue
- ACM Computing Surveys
- Published
- 2023-03-03
- Volume
- 55
- Issue
- 12
- Pages
- 1-38
- ISSN
- 0360-0300 (ISSN-L)
- 1557-7341
- OpenAlex
- W4309674289
- Semantic Scholar
- 3def68bd0f856886d34272840a7f81588f2bc082
- Cited by
- 5,025
- References
- 111
- Citations by year
- 2026: 2296 · 2025: 1441 · 2024: 817 · 2023: 453
- FWCI
- 224.36
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- ACM Computing Surveys (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- Rare & Special e-Zone (The Hong Kong University of Science and Technology) (submittedVersion)
- arXiv (Cornell University) (cc-by)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Speech and dialogue systems
- Keywords
- hallucinations, natural language generation, hallucination mitigation, abstractive summarization, dialogue generation, generative question answering, data-to-text generation, machine translation, large language models, transformer models
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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-09-30, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-09-30 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
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
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.




