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 4,101–4,200 of 4,844
Li, A., Yao, S.-T., Chen, B.-X., Li, X. X., & Guo, S.-Y. (2026) · Frontiers in Public Health · doi:10.3389/fpubh.2026.1909370
“First, LLMs are prone to generating factually incorrect or clinically misleading content, a phenomenon often referred to as hallucination, which in the context of depression psychoeducation could inadvertently reinforce maladaptive beliefs or provide unsafe advice (21).”
Wu, J., Gan, L., Zhang, X., & Shu, Y. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.7193839
Mun, J., Kim, N., & Choi, J. (2026) · Electronics · doi:10.3390/electronics15153307 · Cited by 1
Palermi, S., Saglietto, A., D’Ascenzo, F., Ferrari, G. M. D., Bruno, F., Tse, G., Kizilkilic, S. E., Tjong, F. V. Y., Ruipérez-Campillo, S., Keser, N., et al. (2026) · Progress in Cardiovascular Diseases · doi:10.1016/j.pcad.2026.07.005 · Cited by 2
Constraint-aware specialized large language models for RF circuit reasoning
Chen, J., Liu, J., Wang, Y.-X., et al. (2026) · Microelectronics Journal · doi:10.1016/j.mejo.2026.107404
The local default: Children’s reports of parental birthplace track population base rates
Eriksson, K. (2026) · Judgment and Decision Making · doi:10.1017/jdm.2026.10044
Evaluation of Plate Homogeneity in Cell-Based Potency Assays Using Large Language Models
Košir, R., Uzar, A., Brvar, L., & Oven, I. (2026) · Biophysica · doi:10.3390/biophysica6040066
Grounded Refusal in Dialogue Troubleshooting Agents via Constrained Reinforcement Learning
Li, X., Jin, D., Chen, S., Li, W., & Guo, J. (2026) · International Conference on Computer Intelligence and Software Engineering (CICSE) · doi:10.1109/cicse70286.2026.11620948
“Truthfulness and hallucination studies further indicate that fluent answers can still be unsupported or misleading when evidence is absent [25], [26].”
Li, X., Zh, L., Jiang, S., Chen, J., & Chen, W. (2026) · International Journal of STEM Education · doi:10.1186/s40594-026-00641-y · Cited by 1
“In contrast, the LLM-based agent can further expand its intelligence capabilities by acting in different roles and invoking external tools to tackle complex problems (Ji et al., 2023; Wang et al., 2024).”
Yang, J., Wang, R., Guo, F., & Chen, Y. (2026) · Frontiers in Medicine · doi:10.3389/fmed.2026.1893231 · Cited by 1
“To move beyond aggregate metrics and analyze model failures more systematically, we developed a four-category error typology informed by prior work on hallucination in natural language generation (10).”
Wang, L., & Poenaru, D. (2026) · medRxiv · doi:10.64898/2026.07.23.26358815
Automating Legal Tasks with AI and Other Technologies
Branting, L. K., & Bonardi, M. (2026) · Cambridge University Press eBooks · doi:10.1017/9781009779487.008
“Who is Driving Anyway?” A Critical Survey of Trust and Confusion in “Symbiotic” Teleoperation
Haddad, S., Zouinar, M., Mualla, Y., & Abbas‐Turki, A. (2026) · International Conference on Human-AI Interaction and Experience Design (HAXD) · doi:10.1109/haxd70072.2026.11621021 · Cited by 1
Knowledge Organization of Local Standards Using LLMs and Knowledge Graphs
Zhu, L., Rahman, S. A., & Timan, H. (2026) · International Conference on Artificial Intelligence, Big Data and Algorithms (CAIBDA) · doi:10.1109/caibda70336.2026.11621066
“General-purpose LLMs may generate plausible but unverifiable content, while domain-specific normative documents require schema control, source traceability, and expert verification [7,8].”
InfluentialChansanam, W., Detthamrong, U., & Li, C. (2026) · Iberoamerican Journal of Science Measurement and Communication · doi:10.47909/ijsmc.382
“However, these models are susceptible to a well-documented failure mode, namely hallucination, in which fluent yet factually incorrect statements are generated with a high apparent confidence (Ji et al., 2023).”
Gao, W., & Qin, Y. (2026) · International Conference on Artificial Intelligence, Big Data and Algorithms (CAIBDA) · doi:10.1109/caibda70336.2026.11621390
“In case they are used in tasks requiring knowledge, they can be vulnerable to problems like factual inconsistency, lack of coverage of domain knowledge, and hallucination creation [2 - 3].”
Prompt Engineering Techniques for Enhanced LLM Reasoning Capabilities
Shah, J. A., Jangid, H., Jangid, H., Patel, M., Gadiraju, R. K., & Mahmud, F. (2026) · International Conference on Intelligent Technologies (CONIT) · doi:10.1109/conit69683.2026.11621384
“A major obstacle to deploying LLMs is hallucination, where models produce plausible but incorrect or inconsistent responses [4].”
Contamination Percolation in Multi-Agent LLM Systems: A Measurement Framework and Benchmark
Sharma, A. K. (2026) · IEEE Access · doi:10.1109/access.2026.3717722
“Hallucination [16] is the generation of content that is unsupported by, or inconsistent with, the model’s input or world knowledge.”
InfluentialHow Information Detail and Confidence in AI Responses Shape User Beliefs
Wu, Z., Jobanputra, M., Demberg, V., Hullman, J., & Feit, A. M. (2026) · International Conference on Human-AI Interaction and Experience Design (HAXD) · doi:10.1109/haxd70072.2026.11620844
“Besides persuasion, more recent studies examined the broader potential of LLMs to influence users at higher cognitive levels such as manipulation [12], [13] and deception [8], [14].”
From Hallucination Neurons to Role-Structured Motifs in LLMs
You, J., Chen, J., Xu, Z., & Gao, Y. (2026) · International Conference on Mechatronics, Robotics, and Artificial Intelligence (MRAI) · doi:10.1109/mrai70020.2026.11621705
“Existing mitigation efforts have largely focused on training objectives, retrieval augmentation, calibration, and decoding control[2][3][4].”
Large Model Citation Illusion Detection Based on Multimodal Information Fusion
Yin, L., & Yin, S. (2026) · International Conference on Artificial Intelligence, Big Data and Algorithms (CAIBDA) · doi:10.1109/caibda70336.2026.11621108
“Large models often fabricate references that do not exist, or tamper with the titles and conclusions of real documents, mislabel the author's year and journal information, and may cite documents unrelated to the subject [1].”
Duong, N.-A.-T., Nguyen, A., & Huynh, L. (2026) · Behaviour and Information Technology · doi:10.1080/0144929x.2026.2706668
Westermann, H., & Giacalone, M. (2026) · Cambridge University Press eBooks · doi:10.1017/9781009779487.009
Amoei, M., Fraser, K., & Poenaru, D. (2026) · doi:10.2196/preprints.107982 · Cited by 1
Editorial: Explainability and transparency of artificial intelligence: emergence of AI forensics
Goel, S., & Tejay, G. P. S. (2026) · Organizational Cybersecurity Journal Practice Process and People · doi:10.1108/ocj-03-2026-039
Sun, C., Fang, J., Ni, L., Yu, K., Lu, Y., & Lu, Y. (2026) · Information Processing & Management · doi:10.1016/j.ipm.2026.105071 · Cited by 1
S-AI-Reasoning: Reasoning as Regulated CanonicalizationPreprint
Slaoui, S. C. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6888738 · Cited by 5
KGMedQA: An automated and comprehensive KG-based benchmark for biomedical LLM assessment
Hao, Q., Cheng, K., Liu, X., Wang, H., Shao, J., & Yang, C. (2026) · AI Open · doi:10.1016/j.aiopen.2026.07.001
Intelligent Automation Improved Efficiency in Pharmacovigilance Safety Signal Assessment
Warner, J. B., Souza‐Teodoro, L. H., Jardim, A. P., & Albera, C. (2026) · Clinical Pharmacology & Therapeutics · doi:10.1002/cpt.70409 · Cited by 1
“15 Indeed, AI performance varied across case elements; the lowest F1 score (0.444, accuracy 94.32”
InfluentialQuest of Aivengarde: Comparative Study of Player Experience Across LLM Dialogue Systems
Rimer, E., Petersen, A. K. S., Ploug, R., & Scirea, M. (2026) · International Conference on the Foundations of Digital Games (FDG) · doi:10.1145/3815598.3815622
Research on Evidence-Relation Constraint Methods for Hallucinations in Sentiment Analysis
Si, F. (2026) · International Conference on Big Data & Artificial Intelligence & Software Engineering (ICBASE) · doi:10.1109/icbase70763.2026.11619506
“Hallucination research [18,19,20] shows that generated outputs can conflict with facts, sources, or context.”
Bazi, I. E., Ezzine, J., & Ghazoui, B. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-2600-0413-5.ch006
Husk, S. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6884282
Peças, P., Pires, D., & Jorge, D. P. (2026) · International Journal of Mathematical Engineering and Management Sciences · doi:10.33889/ijmems.2026.11.4.062
A compositional dual-hand action recognition method for LLM-driven assembly assistance
Zheng, H., Huang, J., Xue, J., Li, X., & Xu, X. W. (2026) · Journal of Manufacturing Systems · doi:10.1016/j.jmsy.2026.07.017
Complexity-aware synergistic distillation for efficient abstractive summarization
Wang, G., Wang, D., Liu, S., Wang, K., Liu, X., Qiang, J., & Liu, J. (2026) · Knowledge-Based Systems · doi:10.1016/j.knosys.2026.116782
Ammo, T., Englich, M., Adrego, F. D. S., Jacobi, M., Wilckens, A., Kruppa, P., Bonaventura, B., Niemuth, S. S., Weiler, S. M., Wachenfeld-Teschner, V., et al. (2026) · Research Square · doi:10.21203/rs.3.rs-10035573/v1
Nath, S., Basu, A., Dasgupta, S., & Das, S. (2025) · Indian Conference on Computer Vision, Graphics and Image Processing · doi:10.1145/3774521.3774542 · Cited by 8
“Hallucination degrades the model performance and greatly harm the user experiences in real-world applications [14].”
Retrieval-Augmented Generation for Trustworthy Enterprise LLM Assistants
B, B. P. (2026) · International Journal of Information Technology Research Studies (IJITRS) · doi:10.63090/ijitrs/3139.3209.0033
Yan, W. (2026) · IEEE Access · doi:10.1109/access.2026.3718514
“Existing hallucination mitigation methods include contrastive decoding, quality estimation, semantic similarity checking, uncertainty estimation, and post-editing [19].”
Cao, Y., Tang, Y., Deng, L., Liu, X., Feng, Z., & Gao, Y. (2026) · Journal of Advanced Nursing · doi:10.1111/jan.70691 · Cited by 1
Rizzo, A. ”., Mozgai, S., Hale, B., & Shilling, R. D. (2026) · Elsevier eBooks · doi:10.1016/b978-0-443-33151-0.00003-4 · Cited by 2
LLM-Based Detection of Tangled Code Changes for Higher-Quality Method-Level Bug Datasets
Opu, M. N. I., Wang, S., & Chowdhury, S. (2026) · International Conference on Mining Software Repositories (MSR) · doi:10.1145/3793302.3793351 · Cited by 9
Başkan, B., Evcimen, Y., Timuçin, Ö. B., Yarımağa, İ., & Sağmış, C. (2026) · Muş Alparslan Üniversitesi Sağlık Bilimleri Dergisi · doi:10.71255/maunsbd.1913060
Rosita, Y. D., Hakim, A. R., & Satwika, Y. W. (2026) · International Conference on Smart Computing, IoT and Machine Learning (SIML) · doi:10.1109/siml69834.2026.11621191
“…of speech-based interaction and conversational AI specifically for structured self-counseling aimed at preliminary psychological insight, particularly within educational settings The Gemini API provides advanced capabilities in contextual understanding and adaptive dialogue generation [10].”
Hallucination Is Not One Thing: A Two-Axis Taxonomy for Structured Diagnosis in Generative AI
Khan, M., Billion-Polak, P., & Khoshgoftaar, T. M. (2026) · IEEE Access · doi:10.1109/access.2026.3718830
Neuro-Symbolic Instruction Tuning for Explainable Mahjong Agents via Two-Stage Dual-LoRA
Fang, Z., Xu, J., Yu, J., Li, H., Chen, S., Li, J., & Yoshioka, M. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-31673-8_11
Xue, J., Tang, Y., Mao, Y., Wang, Z., Pan, Z., Nie, B., & Qi, R. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-31404-8_7
The Impact and Potential of Large Language Models in Personalized Learning Environments
Aderghal, Y., Ghazi, A. E., & Hafidi, H. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-23844-3_30
SERC: LDPC-Inspired Semantic Error Correction for Retrieval-Augmented Generation
Kim, G., Park, J., Kim, J., Han, S., Son, K., & Jang, I. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-31920-3_25
Kim, S. J., Ahn, D., Kim, H. R., Kim, S., Kim, K., & Ko, B. C. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-31452-9_44
Alsayed, M. I., Alsayed, M. M., Maghrabi, A., Alzahrani, A., Felemban, A., Badirah, S., Alabdulkarim, F. M., Sheitt, H., Hashem, D., Meer, E., et al. (2026) · International Journal of Medical Informatics · doi:10.1016/j.ijmedinf.2026.106637
URAG 2.0: An Agentic Dual Retrieval Framework for Enhanced Reasoning in RAG-Based QA Systems
Nguyen, L., Vo, Q. N. T., Nguyen, T. T., & Quan, T. (2026) · Communications in computer and information science · doi:10.1007/978-981-92-2584-2_37
Koscelny, S. N., Sadralashrafi, S., & Neyens, D. M. (2026) · Computers in Human Behavior Artificial Humans · doi:10.1016/j.chbah.2026.100370
Li, X., & Liu, P. (2026) · Neurocomputing · doi:10.1016/j.neucom.2026.135011
Bailo, D., Lauciani, V., Stefano, R. D., & Bernardi, F. (2026) · Earth Science Informatics · doi:10.1007/s12145-026-02197-5 · Cited by 1
“Consequently, they can produce responses that sound plausible but may be incorrect or unsubstantiated–a phenomenon often described as "hallucination" (Ji et al. 2023).”
Cheng, J. (2026) · Behavioral Sciences · doi:10.3390/bs16081316
“Consistent with this view, a perspective paper https://doi.org/10.3390/bs16081316 on LLMs in mental health applications stresses that human counselors’ empathy, contextual judgment, and nuanced interpretation remain irreplaceable in high-stakes emotional settings (Ji et al., 2023).”
InfluentialLeite, M. V., Abe, J. M., dos Santos, J. G. B., Thomas, L. C. D. S., & Souza, M. L. H. (2026) · Advances in sustainability science and technology · doi:10.1007/978-3-032-25267-8_12
Knowledge Graph Adapter-Based Augmentation Testbed for Large Language Models
Ali, U., Lynden, S. J., Matono, A., & Amagasa, T. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-34374-1_3
A New Paradigm: Agentic AI for Scientific Discovery
Taktakidze, A. (2026) · Longevity Horizon · doi:10.65649/ukh4n33x · Cited by 1
Kim, T., Kwon, O., Bae, S., & Lee, K. Y. (2026) · Behaviour and Information Technology · doi:10.1080/0144929x.2026.2711036
Enhancing Industrial Fault Diagnosis: A Probabilistic Expert System With LLM-Augmented Validation
Li, Y., Tang, X., Dai, Y., Gao, H., Qian, L., & Zang, Z. (2026) · IEEE Access · doi:10.1109/access.2026.3719708
“LLMs can produce answers that sound convincing yet get the physics wrong—so-called hallucinations—and they lack the physical grounding that factory-floor control systems depend on [12].”
Dialectical systems framework for LLM applications in intelligent construction and operation
Xu, X., & Li, Z. (2026) · Automation in Construction · doi:10.1016/j.autcon.2026.107186
郝洪泽, & Wang, C.-C. (2026) · Journal of theoretical and applied electronic commerce research · doi:10.3390/jtaer21080245
Interval-Valued Epistemic Reasoning for Hallucination-Aware Language Models
Seth, T., & Muhuri, P. K. (2026) · Proceedings of ... IEEE International Conference on Fuzzy Systems · doi:10.1109/fuzz69877.2026.11626347
“Despite their impressive fluency and coherence, LLMs tend to generate factually incorrect or unsupported text, a phenomenon commonly referred to as hallucination [3], [4].”
Effect of evaluation prompt strategies on LLM-as-a-judge reliability in critical care
Yan, J.-Y., Chan, W.-S., Chiu, C., Chao, A., Hsiao, P., Ku, S., & Yeh, Y. (2026) · Anaesthesiology Intensive Therapy · doi:10.5114/ait/221943
IKEA: Intelligent Knowledge Extraction with Reasonable and Agentic AI Agents
Alghamdi, M. M. M., & Alamri, W. A. (2026) · Applied Sciences · doi:10.3390/app16157704 · Cited by 1
Husk, S. (2026) · AI and Ethics · doi:10.1007/s43681-026-01300-2
“Models are reasonably well-calibrated on multiple-choice and true/false formats but struggle on novel task types [ 27].”
Hu, Y., Zhang, W., Feng, F., et al. (2026) · doi:10.1109/rfat69041.2026.11634074
“Many questions require exact values or mechanisms found in only a few papers, making plausible-but-unsupported generations risky—a hallmark of hallucination in neural generation [8].”
Retrieval-augmented generation for medical question answering: a multi-metric performance evaluation
Kökver, Y. (2026) · PeerJ Computer Science · doi:10.7717/peerj-cs.4053
A Multi-Agent Collaborative and Controllable Reasoning Approach for Ship Building Process
Li, N., Yang, G., Cai, Y., Chen, C., Wang, T., & Cheng, L. (2026) · Youth Academic Annual Conference of Chinese Association of Automation (YAC) · doi:10.1109/yac71005.2026.11616153
“…(1) Missing evidence : the model may output confident conclusions without citing verifiable source text, making review and accountability difficult [2], [3]; (2) Long documents and noise sensitivity : technical manuals are lengthy and include substantial irrelevant content, which may cause…”
相引, 眞., Feng, Y., Wang, H. H., & An, W. (2026) · Information Processing & Management · doi:10.1016/j.ipm.2026.105075 · Cited by 5
Elkersh, N. M., Atteya, A. M. A., & Shararah, A. E. A. (2026) · Dentomaxillofacial Radiology · doi:10.1093/dmfr/twag057
A survey on mathematical reasoning and optimization with large language models
Forootani, A., Aliabadi, D. E., & Thrän, D. (2026) · Intelligent Systems with Applications · doi:10.1016/j.iswa.2026.200712 · Cited by 32
Análise de Falhas na Detecção de Alucinação em Textos Jurídico-Policiais em Português
Barcelar, R. R., & Ventura, T. M. (2026) · doi:10.5753/pesquisanuvem.2026.23945
“Estudos como os de Ji et al. (2022) reforçam que a ausência de mecanismos explícitos de grounding contribui para inconsistências factuais, especialmente em tarefas que demandam precisão terminológica e aderência a evidências documentais.”
Ayadi, M. G. (2026) · Network Modeling Analysis in Health Informatics and Bioinformatics · doi:10.1007/s13721-026-00843-y
“Ji et al. (2023 ) comprehensively surveyed hallucina - tion phenomena in natural language generation, categorizing hallucination types: (1) input-conflicting (generating content inconsistent with source document), (2) context-conflicting (generating content inconsistent with prior gen - erated…”
The Non-Fakeness of Fake News: when alleged misinformation turns out to be true
Badawy, W. (2026) · AI and Ethics · doi:10.1007/s43681-026-01289-8
“Hallucination is especially problematic in verification settings because a plausible explanation may be mistaken for evidence [ 16].”
XunZi, an AI biologist, reveals disease-modifying targets
Huang, X., Qin, J., Tang, F., Yang, C., Xu, L., Wei, J., Liu, D., Chen, K., Zhang, C., Chen, M., et al. (2026) · Nature Biomedical Engineering · doi:10.1038/s41551-026-01769-6
Carvalho, D. B. F., Souza, E. D. P., Dias, B. P. M., Detomi, G. H. A., Sousa, P. V. F., Dias, W. M., & Viana, M. C. (2026) · Anais do Workshop sobre as Implicações da Computação na Sociedade (WICS) · doi:10.5753/wics.2026.21040
“…linguísticos, sem compromisso condizente à responsabilidade contextual, o que os torna suscetíveis à geração de informações incorretas ou inventadas ( i.e. , alucinações) e à reprodução de vieses presentes nos dados de treinamento (Bender et al., 2021; Ji et al. , 2023; Weidinger et al. , 2021).”
Singla, P., Singh, J., & Chhabra, G. S. (2026) · Knowledge-Based Systems · doi:10.1016/j.knosys.2026.116785
Khan, S., Samuelsson, J. G., Cai, X., Natarajan, K., & Madhavan, S. (2026) · Drug Discovery Today · doi:10.1016/j.drudis.2026.104754 · Cited by 1
From Command to Conversation: Integrating Agentic AI Into Command-Line Interfaces
Shrivastava, M. (2026) · Computer · doi:10.1109/mc.2026.3673219
Consensus-Gated Execution: A Multi-Agent LLM Architecture for Autonomous Cryptocurrency Trading
Bajgora, A., Kulakov, A., & Merovcı, F. (2026) · Electronics · doi:10.3390/electronics15153453 · Cited by 1
Meridian: In-Memory Acceleration for RAG with Document Attention Decomposition
Liu, C., Huang, Y., Liu, H., Zhang, Y., Qihang, Q. I. U., Li, X., Zheng, L., Liao, X., Jin, H., & Xue, J. (2026) · Proceedings - International Symposium on Computer Architecture/Proceedings · doi:10.1109/isca66397.2026.00041 · Cited by 1
“Large Language Models (LLMs) have achieved remarkable success across diverse domains [6], [59], but their closed-world design makes them prone to hallucinations [24] and knowledge staleness [35].”
Teacher Mediation and the Contingent Promise of Generative AI in Mathematics Education
Canonigo, A. M. (2026) · Journal of Computer Assisted Learning · doi:10.1002/jcal.70307 · Cited by 2
Uncertainty-Aware Structural Attention Features for Hallucination Detection in LLMs
Lokesh, R., Mokshagna, C., Dinesh, K., & Usha, P. (2026) · International Conference on Signal Processing and Communications (SPCOM) · doi:10.1109/spcom71105.2026.11622967 · Cited by 1
“Hallucinations occur across different tasks and model families, representing a significant barrier to the safe deployment of LLMs [8], and detecting them has become an important research problem [18], [19].”
Trust-Guided Multimodal LLM Integration with Reinforcement Learning for Autonomous Driving
Chennaka, S., & Nidamanuri, J. (2026) · IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW) · doi:10.1109/wacvw68408.2026.00192
“…challenges: output reliability varies significantly by scenario (ranging from 95% accuracy in clear weather to 50% in adverse conditions), inference latency (50-100 ms) conflicts with real-time control requirements (10 ms cycle), and hallucinations can propagate to safety-critical decisions [4].”
Marya, A., & Selvaraj, S. (2026) · International Dental Journal · doi:10.1016/j.identj.2026.109802 · Cited by 1
Skill-Based Autonomous Agents for Material Creep Database Construction
Wu, Y., Su, T., Hu, S., & Pan, D. (2026) · Applied Sciences · doi:10.3390/app16157819 · Cited by 1
Shi, J., Ma, Y., Liu, Y., & Shi, B. (2026) · Frontiers in Public Health · doi:10.3389/fpubh.2026.1904062 · Cited by 1
Automated Text-to-SWRL Rule Generation for Tolerance Specification Knowledge Graphs with LLM and RAG
Huang, Z., Luo, J., Fu, H., Kong, C., Zhang, Z., Hu, X., Yang, S., & Cao, Y. (2026) · Procedia CIRP · doi:10.1016/j.procir.2026.03.135
Song, J., Makhdzir, N., Haron, Z., Qin, C., & Wang, L. (2026) · Frontiers in Public Health · doi:10.3389/fpubh.2026.1902424 · Cited by 1
“Beyond the typical concerns about the quality of sources, the trustworthiness issues with generative AI include fabricated citations, untraceable evidence, expressions of overconfidence, delays in model updates, and inconsistent responses to the same query (41–43).”
Cortex: A Hybrid KG-LLM Approach for Multidocument Question Answering
Zawad, N., & Khan, L. R. (2026) · Communications in computer and information science · doi:10.1007/978-981-92-1546-1_40
Spectral radius as a hallucination indicator: Mathematical foundations for AI reliability
Mlaiki, N. M. (2026) · Knowledge-Based Systems · doi:10.1016/j.knosys.2026.116767 · Cited by 1
Tian, H.-C., & Xiao-hui, L. (2026) · Frontiers in Psychology · doi:10.3389/fpsyg.2026.1828685
Jakubski, K. J. (2026) · Cybersecurity & Cybercrime · doi:10.5604/01.3001.0055.7867
Qian, J., Liang, X., Que, L., Pang, Z., Chen, S., & Zhao, F. (2026) · IEEE Access · doi:10.1109/access.2026.3721388
“Nevertheless, directly using LLMs as dispatch assistants remains risky because general models may lack reliable grid-specific knowledge, may hallucinate, and may not satisfy real-time latency, traceability, and authorization requirements [32].”
AgentsKG: A Hierarchical Multi-Agent Framework for Open-Domain Knowledge Graph Construction
Liu, S., Liu, Y., Liu, J., Guo, X., Li, T., Peng, Q., & Wang, W. (2026) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3770855.3818058
Boosting Multimodal Chain of Thought Reasoning by Selective Mixture of Experts
Li, Q., Sun, S., Li, D., Hospedales, T. M., & Gong, S. (2026) · Pattern Recognition · doi:10.1016/j.patcog.2026.114570
PACE: Unleashing the Power of Code Embeddings to Boost AutoML Agents
Zhao, G., Liang, H., Tang, H., Ma, Y., Liu, J., Du, Z., Zheng, Y., Xiao, C., & Hao, J. (2026) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3770855.3817742
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.




