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 1,901–2,000 of 4,844
Hörtling, S., Bause, K., & Albers, A. (2025) · SAE technical papers on CD-ROM/SAE technical paper series · doi:10.4271/2025-01-0289 · Cited by 2
AggTruth: Contextual Hallucination Detection Using Aggregated Attention Scores in LLMs
Matys, P., Eliasz, J., Kiełczyński, K., Langner, M., Ferdinan, T., Kocoń, J., & Kazienko, P. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-97570-7_18 · Cited by 5
“Although definitions of hallucination may vary across sources, the one we adopt defines it as the phenomenon of producing responses that are non-sensical or unfaithful to the provided source content [8].”
Un-trusting the Chat: Designing for Calibrated Trust in Retrieval-Augmented Conversations
von Reeken, T., Salous, M., & Abdenebaoui, L. (2025) · ACM Conference on Conversational User Interfaces (CUI) · doi:10.1145/3719160.3737620 · Cited by 5
Surfacing citizens’ policy perspectives at scale in the age of large language models
Lopez-Lopez, E., & Herzog, S. M. (2025) · Behavioral Science & Policy · doi:10.1177/23794607251346311 · Cited by 4
“62 Human validation of insights derived through such LLM-based approaches is essential in this process.”
Information Retrieval in the Age of Generative AI: The RGB Model
Garetto, M., Cornacchia, A., Galante, F., Leonardi, E., Nordio, A., & Tarable, A. (2025) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3726302.3730008 · Cited by 9
An Empirical Study of Evaluating Long-form Question Answering
Xian, N., Fan, Y., Zhang, R., de Rijke, M., & Guo, J. (2025) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3726302.3729895 · Cited by 7
Iterative attack-and-defend framework for improving TCR-epitope binding prediction models
Zhang, P., Mei, H., Bang, S., & Lee, H. (2025) · Bioinformatics · doi:10.1093/bioinformatics/btaf224 · Cited by 2
“RLAIF helps to reduce hallucinations, seem plausible but are factually incorrect outputs, in large language models (LLMs) by leveraging feedbacks from reward models (Ji et al. 2023).”
Pilgram, L., Kababji, S. E., Liu, D., & Emam, K. E. (2025) · Journal of Medical Internet Research · doi:10.2196/77893 · Cited by 7
“This is because faithfulness is evaluated based on input data adherence, while factfulness requires an external ground truth, making its assessment more challenging [12].”
InfluentialUsing LLMs for Security Advisory Investigations: How Far are We?
Abdullah, B. F., Nugroho, Y. S., Reid, B., Kula, R. G., Shimari, K., & Matsumoto, K. (2025) · International Conference on Smart Computing, IoT and Machine Learning (SIML) · doi:10.1109/siml65326.2025.11080876 · Cited by 7
“While LLMs show promise in vulnerability detection [9] and automated security bug repair [10], their tendency to hallucinate information [11] raises risks.”
None of the above: comparing scenarios for answerability detection in question answering systems
Reyes-Montesinos, J., Rodrigo, Á., & Peñas, A. (2025) · Applied Intelligence · doi:10.1007/s10489-025-06765-y · Cited by 6
From Static to Intelligent: Evolving SaaS Pricing with LLMs
Cavero, F. J., Alonso, J. C. G., & Ruiz–Cortés, A. (2025) · Lecture notes in computer science · doi:10.1007/978-981-96-7423-7_12 · Cited by 4
“One prominent issue is hallucination , where models produce text that appears plausible but is factually incorrect or fabricated [13].”
On the Convergence of Large Language Model Optimizer for Black-Box Network Management
Lee, H., Zhou, W., Debbah, M. A., & Lee, I. (2025) · IEEE Transactions on Communications · doi:10.1109/tcomm.2025.3592598 · Cited by 9
“These prompt engineering methods suffer from the uncertainty in LLM responses, exemplified by the hallucination issue [45], [46], where LLMs generate unfaithful or nonsensi-cal content.”
Incorporating RAG for Factual Hallucination Detection Modeling in Intelligent Educational system
Zheng, L., Ye, J., Zhao, G. L., Luo, S., Nan, M., & Xie, Y. (2025) · International Conference on Computer Science and Technologies in Education (CSTE) · doi:10.1109/cste64638.2025.11092000 · Cited by 4
“…current generative language models are based on a large number of parameters to store knowledge, and there is a risk of over-generalization on untrained data, which is prone to generating some fictional "facts", thus triggering the phenomenon of "factual hallucination" and misleading the users[8].”
A comprehensive study on factual consistency in LLMs for data-to-text generation
Mahapatra, J., & Garain, U. (2025) · Sādhanā · doi:10.1007/s12046-025-02860-5 · Cited by 1
“Although LLMs [25] often achieve higher rankings than earlier DTG models in readability (addressing concerns with fluency and coherence) and informativeness (maintaining content preservation), they continue to struggle with generating factually consistent content [26, 27].”
Romanyukha, A. A., Mazloumi, M., Waelheyns, T. D., Mishra, N., Jacobs, J., & Fitousi, N. T. (2025) · Physica Medica · doi:10.1016/j.ejmp.2025.105090 · Cited by 2
Enhancing robustness of LLM-driven multi-agent systems through randomized smoothing
Hu, J., Dong, Y., Ding, Z., & Huang, X. (2025) · Chinese Journal of Aeronautics · doi:10.1016/j.cja.2025.103779 · Cited by 18
Yang, L., Li, Y., Tan, J., & Mao, L. (2025) · PLoS ONE · doi:10.1371/journal.pone.0330258 · Cited by 7
“However, current LLM still face the challenge of hallucination, where generated content lacks domain constraints [13].”
A Lacanian Interpretation of Artificial Intelligence Hallucination
Wang, Y. (2025) · AI & Future Society · doi:10.63802/afs.v1.i1.93 · Cited by 3
Zhang, Y., Song, Y., Pang, Y., Li, S., Jiang, X., Wang, Y., Li, J., Zhang, M., & Wang, D. (2025) · IEEE Open Journal of the Communications Society · doi:10.1109/ojcoms.2025.3608290 · Cited by 2
“Although GPT has the capability to solve simple mathematical deduction problems through its own reasoning, recent researches have shown that using the tools directly can avoid hallucination [32] of using GPT for mathematical derivation and calculations.”
Hu, X., Lin, R., Huang, Y., Wu, X., Gong, Y., & Huang, H. (2025) · Endocrine · doi:10.1007/s12020-025-04421-6 · Cited by 3
Evaluation and incident prevention in an enterprise AI assistant
Maharaj, A. V., Arbour, D. T., Lee, D., Bhattacharya, U., Rao, A. B., Zane, A. V., Feller, A., Qian, K., Rahman, S., & Li, Y. (2025) · AI Magazine · doi:10.1002/aaai.70028 · Cited by 2
Comparação de Modelos de Embeddings e LLMs para Geração Aumentada por Recuperação em Português
Medeiros, L. S. F., & Oliveira, H. (2025) · Anais do Seminário Integrado de Software e Hardware (SEMISH) · doi:10.5753/semish.2025.9027 · Cited by 4
Furukawa, E., Okuhara, T., Okada, H., Nishiie, Y., & Kiuchi, T. (2025) · Health Informatics Journal · doi:10.1177/14604582251381996 · Cited by 3
HEAL-Summ: a lightweight and ethical framework for accessible summarization of health information
Fisher, A., Srinivasan, K. K., Hillier, S. A., & Mago, V. (2025) · Frontiers in Public Health · doi:10.3389/fpubh.2025.1619274 · Cited by 2
“Such inaccuracies can be particularly harmful in sensitive domains such as healthcare, as it can lead to the spread of misinformation (23).”
Redefining digital health interfaces with large language models
Imrie, F., Rauba, P., & van der Schaar, M. (2025) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2025.1623339 · Cited by 12
“While others have discussed LLMs replacing existing clinical tools (1, 2), we envisage LLMs forming a novel interface by enabling dynamic interactions and facilitating deeper engagement with tools and related information, such as explainability, medical papers, and clinical guidelines (3). as…”
A framework to synthetically generate fine-grained hallucinated data
Saxena, A., Aggarwal, D., Badathala, N., & Bhattacharyya, P. (2025) · Language Resources and Evaluation · doi:10.1007/s10579-025-09864-x · Cited by 7
InReAcTable: LLM-powered Interactive Visual Data Story Construction from Tabular Data
Aodeng, G., Li, G., Feng, Y., Chen, Q., Zhang, Y., & Liu, C. H. (2025) · ACM Symposium on User Interface Software and Technology (UIST) · doi:10.1145/3746059.3747719 · Cited by 8
Lin, M., Xu, C., Qu, X., Xu, B., Wang, Y., Zou, R., & Zhang, Y. (2025) · Digital Health · doi:10.1177/20552076251384233 · Cited by 2
“34 Moreover, the application has been shown to provide fake references in the past.”
Wei, G., Zhang, Z., & Urban, C. (2025) · ACM SIGPLAN International Workshop on Language Models and Programming Languages (LMPL) · doi:10.1145/3759425.3763378 · Cited by 4
Large language model enhanced framework for systematic reviews and meta-analyses
Shen, J., Luo, Z., Jia, D., Wang, S., Sun, F., & Wu, J. (2025) · BMJ Digital Health & AI · doi:10.1136/bmjdhai-2025-000017 · Cited by 8
Zou, L., Zhang, P., Jiang, Y.-Q., Wang, X.-W., Yan, X.-J., Wu, J., Jia, Q., Li, W., Cai, Q., Xuan, Z.-R., et al. (2025) · Frontiers in Oncology · doi:10.3389/fonc.2025.1590230 · Cited by 2
Hamidi, C., Badiy, M., Gaou, S., Amounas, F., Azrour, M., Tribak, H., Alnajim, A. M., & Alabdulatif, A. (2025) · Journal of Advances in Information Technology · doi:10.12720/jait.16.10.1430-1441 · Cited by 5
“Despite these advancements, most existing AI-driven assessment solutions either focus on generic question generation or rely solely on Large Language Models (LLMs) without retrieval augmentation, which can lead to hallucinations and factual inaccuracies [7].”
Nath, D., Ghosh, S., Bhattacharya, S., Hiremath, P., & Deb, D. (2025) · Progress in optical science and photonics · doi:10.1007/978-981-95-1561-5_6 · Cited by 2
Hermida, M., Botturi, L., Addiamando, L., Galloni, M., Beretta, C., Cardoso, F., Luceri, L., Bouleimen, A., Giordano, S., & Saad, M. (2025) · MedienPädagogik Zeitschrift für Theorie und Praxis der Medienbildung · doi:10.21240/mpaed/00/2025.11.07.x · Cited by 2
“AI features do not inherently assess the credibility of the information they provide and are susceptible to hallucinations (Ji et al. 2023).”
Kumari, M., Chauhan, R., Jain, R., & Garg, P. (2025) · Information Fusion · doi:10.1016/j.inffus.2025.103902 · Cited by 6
Promoting Generative AI Literacy among Students and Teachers
Soltani, M., Zarzour, H., & Ladjailia, A. (2025) · Signals and communication technology · doi:10.1007/978-3-031-96855-6_12 · Cited by 2
Wei, Z., Wang, S., Rong, X., Liu, X., & Li, H. (2025) · Communications in computer and information science · doi:10.1007/978-981-95-4088-4_32 · Cited by 19
Hierarchical knowledge graph-based QA systems with retrieval-augmented generation
Yang, W.-C., Li, X., Chang, C., & Roy, D. S. (2025) · Enterprise Information Systems · doi:10.1080/17517575.2025.2580477 · Cited by 5
Ariso, J. M., & Bannister, P. (2025) · AI & Society · doi:10.1007/s00146-025-02757-1 · Cited by 9
Oh, H., Salcedo-Vázquez, A. M., Ramírez-Alpizar, I. G., & Domae, Y. (2025) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros60139.2025.11247005 · Cited by 5
“Although the LLM is a useful tool for instant robotic agents, it faces the inherent issue of hallucinations , where its responses may be out of a given context and lack accuracy [8].”
Kabalci, E., Dörterler, M., & Tekerek, A. (2025) · doi:10.1109/intcec65580.2025.11255830 · Cited by 2
Dai, H., Li, Y., Yan, J., & Zhang, Z. (2025) · Conference on Privacy, Security and Trust · doi:10.1109/pst65910.2025.11268838 · Cited by 6
“1) For existing LLM-based AutoPT frameworks, short-sighted planning and hallucinations in task decomposition [18] and command generation limit the effectiveness of LLMs in managing the overall PT process.”
Personalized dialogue generation through knowledge expansion and in-context learning
Wang, Z., Ma, T., Rong, H., & Jia, L. (2025) · Applied Intelligence · doi:10.1007/s10489-025-07030-y · Cited by 4
“However, LLMs may uncontrollably generate content that is inconsistent with the intended persona or even contains hallucinations [6, 7].”
Benchmarking large language models for cell typing in single-cell RNA-Seq
Xiao, T., Hua, D., Wang, Y., Lu, X., & Zhang, C. (2025) · Briefings in Bioinformatics · doi:10.1093/bib/bbaf677 · Cited by 8
“Furthermore, generalist LLMs still have room for improvement in the depth and precision of their biological context understanding, sometimes struggling to capture subtle state differences or annotate rare cell types [61].”
Variational Dehallucination via Retrieval-Augmented Prompt Learning
Lin, H.-Y., Chang, Y.-C., & Chien, J. (2025) · IEEE Transactions on Audio Speech and Language Processing · doi:10.1109/taslpro.2025.3646477 · Cited by 6
“Hallucination in LLMs refers to the instances where the model-generated outputs deviate from factual correctness, introduce unsupported information, or exhibit semantic incoher-ence [8], [18].”
Med-TAMARA: Trust-Aware Multi-Agent Risk Assessment in Medical AI Dialogue
Wu, J.-Y., & Day, M.-Y. (2025) · IEEE International Conference on Information Reuse and Integration (IRI) · doi:10.1109/iri66576.2025.00052 · Cited by 2
“This inherent variability, arising from the probabilistic nature of LLM inference [1], poses significant reliability challenges, particularly under limited data and strict regulatory conditions.”
Hahnfeldt, R., Laukamp, K. R., Schönfeld, M. H., Galldiks, N., Werner, J.-M., Fervers, P., Görtz, L., Gertz, R. J., Nelles, C., Nasirov, N., et al. (2025) · European Journal of Radiology Artificial Intelligence · doi:10.1016/j.ejrai.2025.100054 · Cited by 2
Luo, X., Chen, Y., Tu, Y., & Qiu, W. (2025) · Knowledge-Based Systems · doi:10.1016/j.knosys.2025.114976 · Cited by 4
Rowen: Adaptive Retrieval-Augmented Generation for Hallucination Mitigation in LLMs
Ding, H., Pang, L., Wei, Z., Shen, H., & Cheng, X. (2025) · Annual International ACM SIGIR Conference on Research and Development in Information Retrieval in the Asia Pacific Region (SIGIR-AP) · doi:10.1145/3767695.3769500 · Cited by 90
Fu, Y., Liu, J., Ban, Q., & Zhou, L. (2025) · Cyber-Physical Systems · doi:10.1080/23335777.2025.2599897 · Cited by 3
“On one hand, the output of LLMs is inherently uncertain – the same query may yield structurally different and factually inconsistent [5] answers.”
Ren, R., Xu, Y., Yao, X., & Cole, S. T. (2025) · Current Issues in Tourism · doi:10.1080/13683500.2025.2609218 · Cited by 20
Self-citationTowards Safe and Trustworthy Healthcare AI: Risk Assessment of Medical Dialogue Using LLMs
Tanioka, H. (2025) · Human-Centric Intelligent Systems · doi:10.1007/s44230-025-00131-4 · Cited by 2
The ethics of using artificial intelligence in writing medical research papers
Yu, S., & Hwang, H. (2025) · Kosin Medical Journal · doi:10.7180/kmj.25.141 · Cited by 2
UQA-RAG: Enhancing Urdu Question-Answer Retrieval With Semantic Similarity and LLMs
Khan, T. F., Arslan, M., Alturise, F., Almutairi, A. H., & Raza, M. T. (2026) · IEEE Access · doi:10.1109/access.2025.3649743 · Cited by 2
“The occurrence of hallucination emerges when models produce answers that contradict factual evidence or the available data which represents a significant drawback of LLMs [18].”
Coutinho, M. A., Ashofteh, A., & Alhelaly, Y. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-10721-3_1 · Cited by 4
Sarwatt, D. S., Kulwa, F., Philipo, A. G., Runyoro, A.-A. K., Ning, H., & Ding, J. (2026) · Artificial Intelligence Review · doi:10.1007/s10462-025-11467-5 · Cited by 4
Synergizing linguistic features and transformer networks for detecting AI-generated text
Annepaka, Y., Kumar, P., Poddar, Y., Pakray, P., & Chunka, C. (2026) · Knowledge and Information Systems · doi:10.1007/s10115-025-02637-6 · Cited by 2
Xiong, Y., Ge, J., & Che, L. (2026) · IEEE Transactions on Smart Grid · doi:10.1109/tsg.2026.3654823 · Cited by 2
Federated knowledge retrieval elevates large language model performance on biomedical benchmarks
Joy, J., & Su, A. I. (2026) · GigaScience · doi:10.1093/gigascience/giag007 · Cited by 3
Zhuang, Z., Lee, X., Zhang, A., et al. (2026) · Information and Software Technology · doi:10.1016/j.infsof.2026.108041 · Cited by 2
Gendi, M., & Andritsos, P. (2025) · doi:10.1109/cascon66301.2025.00115 · Cited by 2
Yutong, G. U. O., & Yang, C. (2026) · Metals · doi:10.3390/met16020162 · Cited by 5
“Hallucination is a measurable failure mode: surveys define it as output that is unfaithful to underlying sources or inconsistent with verifiable facts and summarize detection and mitigation approaches across tasks [95].”
Why Compromise Privacy? Local LLMs Rival Commercial LLMs in Qualitative Analysis
Adeseye, A., Isoaho, J., Virtanen, S., & Mohammad, T. (2025) · Computing, Communications and IoT Applications (ComComAp) · doi:10.1109/comcomap68359.2025.11353130 · Cited by 3
“Prompt design plays an important and decisive role in controlling hallucinations for both local and commercial LLMs [18].”
Lin, S., shao, S., Liu, X., & Su, H. (2026) · Information · doi:10.3390/info17020138 · Cited by 2
ŞAHİN, E. G. (2026) · BMC Oral Health · doi:10.1186/s12903-026-07753-5 · Cited by 7
“Despite their usefulness in information retrieval, LLMs may hallucinate (producing incorrect outputs with high confidence) highlighting the need for caution [22].”
Building the ethical AI framework of the future: from philosophy to practice
Catapang, J. K. (2026) · AI and Ethics · doi:10.1007/s43681-026-01003-8 · Cited by 5
Demirpolat, E., Tuncel, M. S., Gurer, F., & Şanlı, N. M. (2026) · Journal of Oncology Pharmacy Practice · doi:10.1177/10781552261418957 · Cited by 4
Odia, J. O. (2026) · Emerald Group Publishing Limited eBooks · doi:10.1108/978-1-80592-391-620261003 · Cited by 3
Hallucinated intelligence? Cognitive and emotional pathways to generative AI reuse
Liu, C., Cong, Y., & Zhang, Z. (2026) · Aslib Journal of Information Management · doi:10.1108/ajim-05-2025-0252 · Cited by 7
Brain-inspired energy efficient technologies for next-generation artificial intelligence
Chiel, H. J., Coggan, J. S., Datta, G., Fellous, J., Nourse, W. R. P., Quinn, R. D., & Thomas, P. J. (2026) · Biological Cybernetics · doi:10.1007/s00422-026-01038-4 · Cited by 7
Guo, Z., & Duan, Y. (2026) · Electronics · doi:10.3390/electronics15050963 · Cited by 2
Iacob, M., Qawas, A., Balasubramaniam, R., Frydrych, A. M., & Kujan, O. (2026) · Journal of Personalized Medicine · doi:10.3390/jpm16030129 · Cited by 2
“Beyond diagnostic applications, emerging evidence suggests that AI platforms may support broader decision support functions, including the assessment of dental drug–drug interactions [24].”
InfluentialGuiding Reinforcement Learning Using Uncertainty-Aware Large Language Models
Shoaeinaeini, M., & Harrison, B. E. (2025) · IEEE International Conference on Trust, Privacy and Security in Intelligent Systems and Applications (TPS-ISA) · doi:10.1109/tps-isa67132.2025.00045 · Cited by 4
“Aleatoric uncertainty is related to observation errors, such as sensor noise, while Although LLMs demonstrate remarkable capabilities and rapid advancements, they frequently produce incorrect information unexpectedly, including hallucinations [42], disinformation [43], or bias [44].”
Rafi, M. M., Krishnamurthy, A., & Balu, A. (2025) · IEEE International Conference on Trust, Privacy and Security in Intelligent Systems and Applications (TPS-ISA) · doi:10.1109/tps-isa67132.2025.00078 · Cited by 2
“A critical challenge for deploying LLMs in high-stakes applications is their tendency to generate plausible but factually incorrect information—a phenomenon known as hallucination [8].”
Han, G. I., Choi, H., & Son, Y.-J. (2026) · BMC Nursing · doi:10.1186/s12912-026-04518-x · Cited by 8
“Future studies should assess the scope and limits of generative AI contributions to nursing efficiency and patient education and self-management, and examine how these effects influence the quality and safety of nursing practice over time [56, 59].”
InfluentialImproving LLM Retrieval with GraphRAG-SF: Semantic Filtering for Graph-Based RAG Systems
Hosseini, M.-P., Shah, A., Miao, C., & Wei, W. (2025) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata66926.2025.11401790 · Cited by 2
Navigating the Blue Nowhere: A Framework for Mapping Validated Adversarial Trajectories
Gilda, S., Martiny, K., Ho, J., et al. (2025) · IEEE ... International Conference on Data Mining workshops · doi:10.1109/icdmw69685.2025.00124 · Cited by 2
“While Large Language Models (LLMs) have transformed this field with their extraction capabilities, they frequently hallucinate convincing but false information [2].”
Shehmir, S., & Kashef, R. F. (2026) · Frontiers in Computer Science · doi:10.3389/fcomp.2026.1735253 · Cited by 2
Performance of latest AI models, RAG, and MCP on lung cancer-related questions
Zhao, X., Yang, M., Tian, K., Jiang, H., Guo, D., Wang, Y., & Du, J. (2026) · Digital Health · doi:10.1177/20552076261427503 · Cited by 4
“12,13 To improve accuracy, technologies such as retrieval-augmented generation (RAG) and Model Context Protocol (MCP) have been developed.”
Ahn, S. H., & Kim, J. (2026) · JMIR Formative Research · doi:10.2196/90273 · Cited by 3
“AI can synthesize quickly, yet hallucinations can produce plausible but medically inaccurate advice, creating unacceptable liability for leaders accountable for patient safety [3,4].”
Efficient and Verified Research Data Extraction with LLM
Serdiukov, A., Dravgelis, V., Smutin, D., Taldaev, A., Ivanov, A. B., Adonin, L., & Muravyov, S. (2026) · Algorithms · doi:10.3390/a19030214 · Cited by 5
“This observation is consistent with the broader literature: hallucination in structured-domain LLM applications often stems from boundary-condition failures rather than wholesale invention [12–14].”
InfluentialTandon, U. (2026) · Frontiers in Digital Health · doi:10.3389/fdgth.2026.1772345 · Cited by 2
“A common and specific concern is hallucinated output generated by LLMs, which seems reasonable but is factually incorrect or contrived (2, 50).”
Large Language Models in Surgery: Promise, Pitfalls, and Practical Use
Denham, D. T., Wang, C. Y., Maric, E., Hinton, L. R., & Heniford, B. T. (2026) · Journal of Abdominal Wall Surgery · doi:10.3389/jaws.2026.16349 · Cited by 3
“PHI entered into non-approved systems may be retained or reproduced in subsequent outputs depending on platform behavior and settings [63].”
Weikang, K., 温川飙, & Yue, L. (2026) · Digital Chinese Medicine · doi:10.1016/j.dcmed.2026.02.005 · Cited by 2
PoetRAT: A poetry retrieval augmented thoughts framework for Chinese classical poetry generation
ZhePing, Y., Ren, Y., Zhang, T., et al. (2026) · Applied Soft Computing · doi:10.1016/j.asoc.2026.115135 · Cited by 2
Vicente-Martínez, P., Soria‐Olivas, E., Esteve-Mompó, I., Sánchez-Montañés, M. A., García-Escrivà, M. Á., & William, E. (2026) · AI · doi:10.3390/ai7040123 · Cited by 7
Generative Artificial Intelligence Models: A Survey
Almuammar, M., Alsultan, K., & Altukhaim, Y. (2026) · Artificial Intelligence Review · doi:10.1007/s10462-026-11546-1 · Cited by 10
“Addressing bias, fairness, and hallucination is also essential for reliable deployment (Peng et al. 2020 ; Ji et al. 2023 ).”
Lightweight query-adaptive RAG framework for knowledge support in smart manufacturing
Zhou, T., Liu, Y., Kumar, M., & Zou, L. (2026) · Journal of Manufacturing Systems · doi:10.1016/j.jmsy.2026.04.016 · Cited by 8
Rafi, M. S., & Khalil, L. (2026) · Frontiers of digital education. · doi:10.1007/s44366-026-0088-9 · Cited by 8
Poudel, A., & Surface, T. (2026) · Discover Civil Engineering · doi:10.1007/s44290-026-00474-2 · Cited by 3
Knowledge graph-enhanced large language models for explainable causal analysis of crane accidents
Zhang, W., Cao, Y., & Zhang, J. (2026) · Reliability Engineering & System Safety · doi:10.1016/j.ress.2026.112859 · Cited by 5
Narrative and dialogue generation for video-games: A systematic mapping
Salmaze, P. H., Pereira, G. D. S., de Góes, L. M. G., Ranieri, C. M., Toledo, C. F. M., Silvestrini, M. M., & Pereira, L. P. D. B. (2026) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2026.115041 · Cited by 5
O’Sullivan, P., Seppälä, S., & Feller, J. (2026) · Journal of Decision System · doi:10.1080/12460125.2026.2669855 · Cited by 2
Hong, J., Rao, P., Wang, W., & Najafizadeh, L. (2026) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2026.3693965 · Cited by 2
“However, these works incorporated cloud-based LLMs, which could introduce variability in response time due to network and server latency [29], and raise concerns about data privacy [30] and reliability [31] in clinical AAC applications.”
Dashti, M., Khosraviani, F., Meyari, A., Amirzade-Iranaq, M. H., Chaurasia, A., Hefzi, D., Ghadimi, N., Tichý, A., Khurshid, Z., & Schwendicke, F. (2026) · International Dental Journal · doi:10.1016/j.identj.2026.109609 · Cited by 7
Breaking the “80-point ceiling”: librarians as facilitators of LLM clean-up engineering
Li, R. (2026) · The Journal of Academic Librarianship · doi:10.1016/j.acalib.2026.103273 · Cited by 3
Muppidi, N. K. R., Divi, V. R., Gulapalli, S., Rachamalla, S., Tatavarthi, S., & Lakshmi, M. A. (2026) · International Conference on Artificial Intelligence, Systems, and Emerging Technologies (ICAISET) · doi:10.1109/icaiset66439.2026.11541762 · Cited by 7
“Such concerns are consistent with broader findings on hallucination, uncertainty, and brittleness in language-model-based systems [5], [10].”
Creely, E. (2026) · AI in Education · doi:10.3390/aieduc2020020 · Cited by 6
Mota, F. B., Cabral, B. P., Pinto, C. D., Braga, L. A. M., Comarú, M. W., & Lopes, R. M. (2026) · Frontiers in Education · doi:10.3389/feduc.2026.1813497 · Cited by 6
Hilkenmeier, F., Stoltenberg, M., & Stierle, C. M. G. (2026) · Systematic Reviews · doi:10.1186/s13643-026-03228-4 · Cited by 8
“In this sense, the task is less affected by performance degradation associated with longer or more complex input contexts [49] and less reliant on the kind of open-ended generation in which hallucinated content is more likely to arise [16, 35, 37].”
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.




