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 3,201–3,300 of 4,844
Li, J., Ahooie, N. N., Paananen, V., Visuri, A., Szabó, D., Kuutila, M., Yatani, K., Mäntylä, M., & Hosio, S. J. (2026) · Behaviour and Information Technology · doi:10.1080/0144929x.2026.2619100
“This can reduce the likelihood of factual errors by ensuring the model’s output is always supported by authoritative information (Ji et al. 2023).”
Younsoo, P., Park, H.-H., & Yi, Y. (2026) · The Journal of Korean Association of Computer Education · doi:10.32431/kace.2026.29.1.002
Mathematical decomposition of prompt engineering in Large Language Model architecture
Jovanović, M., Živanović, M. M., & Aleksić, A. (2025) · Engineering Today · doi:10.5937/engtoday2600002j
Quality-Based Evidence Summarization for Retrieval-Augmented Generation (RAG)
al Masud, A., Fahim, A. A., Azad, M. S., et al. (2025) · 2025 IEEE 7th International Conference on Sustainable Technologies For Industry 5.0 (STI) · doi:10.1109/sti69347.2025.11367610
“A naive pipeline often returns redundant or noisy passages, reducing faithfulness and causing hallucinations [3].”
BERT Gets a Boost: CNN and LSTM Integration for Smarter Wikipedia Question Answering
Wijaya, A. V., Hidayati, S. C., Anggraini, R. N. E., Abousaleh, F. S., & Anistyasari, Y. (2025) · International Conference on Information & Communication Technology and System (ICTS) · doi:10.1109/icts67612.2025.11369633
“However, their practical deployment introduces challenges—most notably, high computational cost, energy usage, latency, and limited interpretability [24], [25].”
Dialogue Meets Data: A Large Language Model Approach for Efficient Linked Data Retrieval
Mussa, O., RANA, O. F., Goossens, B., Orozco‐terWengel, P., & Perera, C. (2026) · ACM Transactions on the Web · doi:10.1145/3796550 · Cited by 2
Zhang, X., Tian, X., & Chen, X. (2026) · Aslib Journal of Information Management · doi:10.1108/ajim-05-2025-0270 · Cited by 1
NordDRG AI Benchmark for Large Language Models
Pitkäranta, T. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-16451-3_14
Smith, K., & Maclin, R. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-15638-9_18
Hybrid Knowledge Base Architecture for Real-Time Conversational AI
Bharathi, A. V., & Nitnaware, P. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-17840-4_3
Ruas, P., Conceição, S. I. R., & Couto, F. M. (2026) · ACM Computing Surveys · doi:10.1145/3796222 · Cited by 2
Talking to Our Students About Generative AI
Rebelsky, W. (2026) · ACM Technical Symposium on Computer Science Education (SIGCSE) · doi:10.1145/3770762.3772642 · Cited by 1
Li, B., Cheng, K., & Kalluri, R. (2026) · Extracellular Vesicle · doi:10.1016/j.vesic.2026.100106
Furukawa, E., Okuhara, T., Suzuki‐Chiba, H., Liu, M., Okada, H., & Kiuchi, T. (2026) · Patient Education and Counseling · doi:10.1016/j.pec.2026.109546 · Cited by 3
Discursive behavior of generative language models in geopolitical and humanitarian contexts
Giacalone, M. (2026) · Discover Artificial Intelligence · doi:10.1007/s44163-026-00965-2 · Cited by 4
Heterogeneous Information-Driven Contrastive Optimization for Abstractive Summarization
Wang, P., Lu, J., & Gong, S. (2026) · Information Systems Frontiers · doi:10.1007/s10796-026-10704-8
“The hallucination problem refers to the generation of plausible-sounding but incorrect or nonsensical informa - tion (Ji et al., 2023 ).”
A Comprehensive Review on Humanoid Robots: Perspectives from Academia and Industry
Li, W., Yang, G., Wu, J., Pan, C., Sheng, L., & Zhang, Q. (2026) · ENGINEERING Information Technology & Electronic Engineering · doi:10.1631/eng.itee.2025.0105 · Cited by 27
SIRAG: Towards Stable and Interpretable RAG with A Process-Supervised Multi-Agent Framework
Wang, J., Wu, Z.-H., Lu, S., Li, Y., & Huang, X. (2025) · 2025 International Conference on Intelligent Robotics and Automatic Control (IRAC) · doi:10.1109/irac67707.2025.11381248 · Cited by 2
“Nevertheless, they still suffer from knowledge staleness and hallucination [4][5], as their internal parameters cannot be continuously updated with external information.”
Sprachmodelle zur Unterstützung des Abweichungsmanagements in der Produktion
Wang, Y., Donecker, M., Zhao, K., Lahuec, M., & Metternich, J. (2026) · Zeitschrift für wirtschaftlichen Fabrikbetrieb · doi:10.1515/zwf-2026-1025
“Fehlende Interpretierbarkeit sowie so-genannte „Halluzinationen“, also von Mo-dellen generierte, aber faktisch unzutref-fende Informationen, begrenzen derzeit das Vertrauen in LLM-basierte Anwendungen [15].”
CAN LARGE LANGUAGE MODELS ACT AS “CO-AUDITORS”?
Emekci, H. (2026) · Denetişim · doi:10.58348/denetisim.1782835
Bridging Fact and Generation: Comparative Insights into Hallucination Management in LLMs
Patil, P., Deshpande, H., Padmawar, S., Kassetty, A. V., & Agarkar, A. A. (2025) · IEEE Pune Section International Conference (PuneCon) · doi:10.1109/punecon67554.2025.11378882
“[9], [13], [14].”
FiR-Rad: Fine-Grained Reinforcement With Structured Reasoning for Chest X-Ray Report Generation
Mei, X., Yang, L., Gao, D., Cai, X., Han, J., & Liu, T. (2026) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2026.3665561 · Cited by 6
“This lack of transparency continues to hinder clinical acceptance and can result in persistent challenges such as factual inaccuracies, information omissions, and textual hallucinations [20], [21].”
WITHDRAWN: From Interpolation to Extrapolation: How Language Models Theorize in Data-Sparse RegimesPreprint
Khaliq, S. (2026) · Research Square · doi:10.21203/rs.3.rs-8875396/v1
Yin, J., Xie, X., Chen, J., Guo, S., & Cui, J. (2026) · Electronics · doi:10.3390/electronics15040870
“Ji et al. [11] subdivided large model hallucinations into two categories: factual hallucinations and faithfulness hallucinations.”
GPT-2++: An Optimized GPT-2 for RAG by Integrating BERT, Prompt Engineering, and Fine-Tuning
Yamout, F., & Hasan, H. A. (2025) · doi:10.1109/fllm67465.2025.11390895
“The core findings from the experiment are: GPT-2++ vs. GPT-2 Base [13]: Our GPT-2++ system overwhelmingly outperformed the base GPT-2 model, achieving a higher score in 906 out of 1,000 [14] queries (90.6%).”
InfluentialFCP-LLM: Functional Coverpoint Plan Generation Using LLM in Early Design Verification Stage
Lin, Z., Guo, Z., Wang, C., Zheng, R., Ji, Y., Lim, Y. W., Zhang, Y. H., Rokhani, F. Z., & Li, Y. (2025) · IEEE Asia Pacific Conference on Circuits and Systems (APCCAS) · doi:10.1109/apccas67402.2025.11377100
“However, LLM models may produce hallucinations when processing complex specifications and HDL code due to limited SoC design knowledge in their training datasets [17].”
Fang, Y., & Zhou, T. (2026) · Information & Management · doi:10.1016/j.im.2026.104322
Democratizing Financial Education: An API-Driven Platform Powered by Retrieval-Augmented Generation
Eleuterio, D. S., Oliveira, P. F., & Matos, P. (2025) · International Conference on Education Technology and Computer · doi:10.1109/icetc66579.2025.11387160
“Standard generative models can be prone to factual inaccuracies or ”hallucinations,” where the model generates plausible but incorrect information [9].”
WITHDRAWN: From Interpolation to Extrapolation: How Language Models Theorize in Data-Sparse RegimesPreprint
Khaliq, S. (2026) · Research Square · doi:10.21203/rs.3.rs-8875396/v2
Research on the Method of Factual Hallucination Relief Based on blockchain Large Language Model
Li, P., Yang, Y., Zhang, J., & Cheng, S. (2025) · International Conference on Blockchain Technology and Information Security (ICBCTIS) · doi:10.1109/icbctis66509.2025.11387485
“If the statement of the question is close to the statement in the training data, the probability of giving the correct answer is higher[13][14].”
Shi, X., Tian, Z., Tan, S., & Wang, X. (2026) · medRxiv · doi:10.64898/2026.04.02.26350091
Mapping the Danish National Health Registries to the OMOP Common Data Model (Preprint)Preprint
Dobreva, J., Yaqoob, M., Alves, F. S., Dahl, S. C., Vesteghem, C., Nielsen, P. B., Hose, K., Bøgsted, M., & Sagi, T. (2026) · doi:10.2196/preprints.93317
Saßmannshausen, T. M., & Wagener, S. (2026) · Qeios · doi:10.32388/gag6kd.2
Trust in Artificial Intelligence Technologies
Кобринский, Б. А. (2025) · Scientific and Technical Information Processing · doi:10.3103/s0147688225700455 · Cited by 1
“This phenomenon is called “artificial hallucination” [44, 45].”
Practical Frameworks for Integrating Artificial Intelligence in Forensic Psychology
Rilen, S. (2026) · Journal of Forensic Psychology Research and Practice · doi:10.1080/24732850.2026.2630383 · Cited by 1
Robustness and Reliability of GenAI Solutions
Pandey, R. K., & Bandapati, G. (2026) · Springer eBooks · doi:10.1007/978-3-032-15606-8_6
LieToMe: A Multi-Agent System for Detecting Hallucinations in Large Language Models
Blitti, K. E. K., Tola, F. G., Shukla, M., Ladva, V., & Dholakia, N. H. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-17834-3_29
MAGNET: Counterfactual samples synthesizing for mitigating hallucination in large language models
Kim, B. S., Kim, B., & Jang, B. (2026) · PLoS ONE · doi:10.1371/journal.pone.0340812 · Cited by 1
“LLM often produces plausible sentences that have no basis in truth [8,16]; this is because LLM learns shortcuts by relying on spurious features when learning, and spurious features include word-overlap, priming, surface form, and co-occurrence [10,17–19].”
Joint Knowledge Graph Reasoning and Large Language Models for Hierarchical Task Planning
Cai, H., Ou, Y., Li, K., & Kong, F. (2025) · IEEE International Conference on Robotics and Biomimetics (ROBIO) · doi:10.1109/robio66223.2025.11375952
“However, the reliability of LLM-generated action sequences remains a major concern, primarily due to inherent issues such as hallucinations [3].”
Quality, Compliance, and Ethical Responsibility
Yu, J. S., & Yao, Y. (2026) · Springer eBooks · doi:10.1007/978-981-95-4632-9_13
Adeseye, A., Isoaho, J., Virtanen, S., & Tahir, M. (2026) · IEEE International Conference on AI in Cybersecurity (ICAIC) · doi:10.1109/icaic67076.2026.11395706
“Studies have shown that unclear and noisy input texts increase the risk of hallucinations [8].”
VQArt-Bench: A Semantically Rich VQA Benchmark for Art and Cultural Heritage
Alfarano, A., Venturoli, L., & del Castillo, D. N. (2025) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw69036.2025.00761 · Cited by 10
“To address this limited expressiveness, subsequent research turned to Neural Question Generation (NQG) models, which promised greater linguistic variety [33] but introduced new challenges, including a critical propensity for ”hallucinating” factually ungrounded content [23].”
Lai, X., Pan, X., Su, Y., & Zhang, G. (2025) · International Conference on Artificial Intelligence, Robotics, and Communication (ICAIRC) · doi:10.1109/icairc68035.2025.11384878
“Large Language Models (LLMs) such as GPT-3 [1], GPT-4 [2], and LLaMA [3] achieve strong performance across NLP tasks but still generate hallucinated or outdated content in knowledge-intensive scenarios [4].”
Retrieval-Augmented Generation
Yu, J. S., & Yao, Y. (2026) · Springer eBooks · doi:10.1007/978-981-95-4632-9_4
Jiang, Y., Li, Y., Li, H., Wu, S., & Feng, Y. (2025) · International Conference on Robotics, Intelligent Control and Artificial Intelligence (RICAI) · doi:10.1109/ricai68060.2025.11385056
“…[5]; long-tail knowledge is sparse in training corpora, resulting in insufficient coverage and reasoning errors in specialized fields such as medicine, law, and traditional opera; moreover, their reasoning mechanisms lack interpretability, making generated content prone to “hal-lucinations” [6].”
Reinforcement Learning With Large Language Models for Long-Horizon Reasoning
Kumar, A., Singhal, P., & Gupta, R. K. (2025) · International Conference on System Modeling & Advancement in Research Trends (SMART) · doi:10.1109/smart66937.2025.11389227
“Any meaningful and responsible adaptation of RL-LLM systems necessitates collaboration among technical researchers, ethicists, and policymakers.”
Unifying Medical Large Language Models and Knowledge Graphs: BioBERT, ICD-10 and the UK Biobank
Oatley, G., Choudhury, T., & Grace, F. (2025) · Health Informatics and Knowledge Management Conference (HIKM) · doi:10.1145/3774816.3774844
Zhang, T., Lyu, J., & Cai, Y. H. (2025) · Asia-Pacific Software Engineering Conference (APSEC) · doi:10.1109/apsec66846.2025.00018
Tractable Asymmetric Verification for Large Language Models via Deterministic Replicability
Chong, Z., Ohsaki, H., & Kuan, B. N. C. (2025) · IEEE Student Conference on Research and Development (SCOReD) · doi:10.1109/scored68498.2025.11399166 · Cited by 1
“Central to this challenge is the lack of efficient, scalable mechanisms for agents to verify the computational integrity of each other’s outputs in LLMs [7].”
KAISA: Knowledge And Internal States Augmented Agent for Chemistry Knowledge Question Answering
Dai, W., Pan, Y., Ge, Z., Li, X., & Zhou, Y. (2026) · International Conference on Consumer Electronics and Computer Engineering (ICCECE) · doi:10.1109/iccece69169.2026.11399756
“As noted by [3], the probabilistic nature of next-token prediction often leads to factual fabrications when the model faces long-tail scientific knowledge or complex calculations that exceed its pre-trained corpus coverage.”
Zhang, Y., Zeng, Y., Zhu, Y., Song, M., Wang, W., & Ou, Z. (2025) · International Conference on Data Science and Intelligent Systems (DSIS) · doi:10.1109/dsis67228.2025.11390158
“Nevertheless, LLMs frequently “halluci-nate” by producing content that is unverifiable or outright false [1] [2], a limitation that severely hampers their trustworthy deployment in high-stakes domains such as healthcare, law, and education [3] [4].”
The Evolution of the Brain From Analog to Digital: The Birth of Artificial Intelligence
Nahass, G. R., & Patel, P. K. (2026) · Journal of Craniofacial Surgery · doi:10.1097/scs.0000000000012364
Large language models show Dunning-Kruger-like effects in multilingual fact-checking
Qazi, I. A., Khan, Z., Ghani, A., Raza, A. A., Qazi, I. A., Sajjad, W., Ali, A., Javaid, A., Sohail, M. A., & Azeemi, A. H. (2026) · Scientific Reports · doi:10.1038/s41598-026-39046-w · Cited by 12
YAN, S., Wang, W., Chen, H., & Wang, J. (2026) · The Journal of Supercomputing · doi:10.1007/s11227-026-08333-8
AI Hallucinations as an Existential Risk to Human Society: Mitigating the Risks
Goltz, N., & Yuxin, Z. (2026) · BENTHAM SCIENCE PUBLISHERS eBooks · doi:10.2174/9798898812256126040007
DAMA: A Dual Alignment Framework for Enhanced LLM-Powered Recommendations
Zeng, Q., Chen, L., Qiu, J., Liu, F., Chen, H., & Chen, L. (2025) · Proceedings · doi:10.1109/icdm65498.2025.00182
“However, in recommendation scenarios, such motivational inference often lacks complete contextual information or explicit supervision, making LLMs prone to generating seemingly coherent yet unverified conclusions in the face of insufficient evidence (Hallucination) [7], [10].”
Harley, B., Syed, F. H., & Ramirez, L. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6304172
Enhancing Medical Question Answering via a Multi-Agent Debate Framework
Lu, D., Jiang, V., Wang, S., Zhu, M., & Zhai, S. (2026) · IEEE Annual Computing and Communication Workshop and Conference (CCWC) · doi:10.1109/ccwc67433.2026.11393824
“However, in complex and cross-domain medical reasoning scenarios where information is often incomplete, current models still exhibit unstable reasoning chains, hallucinations, and premature answering [7], posing significant risks in safety-critical medical applications.”
Critical thinking in human-AI collaborationsPreprint
Bilen, E., & Hervé, J. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6304755 · Cited by 1
TAMO-FoA: Tool-Augmented AIOps for Reliable RCA in Cloud-Native Systems
Mittal, A., Kandi, K., Syed, R., & Mahajan, S. (2026) · IEEE Annual Computing and Communication Workshop and Conference (CCWC) · doi:10.1109/ccwc67433.2026.11393712
“1) Advanced RAG and Verification: Simple retrieval-augmented generation (RAG) [6] can still lead to hallucinations [7] if the LLM ignores or misinterprets the retrieved context.”
Small Models, Big Impact: Tool-Augmented AI Agents for Wireless Network Planning
Zhang, Y., Kishk, M. A., & Alouini, M. (2026) · IEEE Communications Magazine · doi:10.1109/mcom.001.2500402
“…indicates that even advanced LLMs hallucinate technical specifications in 5–15% of domain-specific queries, a rate unacceptable in mission-critical deployments such as wireless network planning, where inaccurate calculations can result in inadequate coverage and costly infrastructure failures [2].”
Ishadya, A. P. D., Withanaarachchi, A. S., & Jayalal, S. (2025) · International Conference on Advances in Technology and Computing (ICATC) · doi:10.1109/icatc68823.2025.11407841 · Cited by 2
“For example, [13] describe scenarios where a LLM-generated medical summary (based on patient records) includes false information that could lead to harmful treatment decisions.”
InfluentialLEAP: Learning Enhanced by AI Personalization for Customized Student Support
Zhang, S., Hussain, S., & Zoha, A. (2025) · International Conference on AI-Driven STEM Education and Learning Technologies (AISTEMEDU) · doi:10.1109/aistemedu67077.2025.11403909
“Existing research define s hallucinations i n LLMs as “mod-els generating untrue/unsupported information in internally consistent language” [4].”
Talebian, S. H., Ibrahim, A. O., & Syed, H. J. (2025) · International Conference on Advances in Machine Intelligence and Cybersecurity Technologies (AMICT) · doi:10.1109/amict65811.2025.11402818 · Cited by 1
“Hallucination and Factual Reliability present significant challenges as models generate plausible but incorrect information, undermining reliability in knowledge-intensive applications [37].”
McOwiti, A., Hao, X., Lin, R. Z., Rasmy, L., Tao, C., & Fenton, S. H. (2026) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2026.105002 · Cited by 1
A text consistency evaluation method combining unified modeling and mixture-of-expert networks
Zhang, L., Xiong, W., Yang, R., & Li, M. (2026) · Expert Systems with Applications · doi:10.1016/j.eswa.2026.131809
HAMZAH, A. B., & Teng, S. (2026) · Research Square · doi:10.21203/rs.3.rs-8929856/v1
SynthEHR-eviction: enhancing eviction SDoH detection with LLM-augmented synthetic EHR data
Yao, Z., Zhao, Y., Mitra, A., Levy, D. A., Druhl, E., Tsai, J., & Yu, H. (2026) · npj Digital Medicine · doi:10.1038/s41746-026-02473-0 · Cited by 3
Reimagining Reality: Innovations in AI Hallucination Management and Trustworthy Generation
Nag, A., & Chaudhuri, D. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-7534-2.ch012 · Cited by 1
Lightweight Retrieval-augmented language model
Li, Y., Zhang, D., Peng, Y., Zhang, H., & Liu, J. (2025) · International Conference on Advanced Control Systems and Automation Technologies (ACSAT) · doi:10.1109/acsat68407.2025.11406960
“Due to their limited parameter scale, these models exhibit insufficient knowledge storage and representation capabilities, making them prone to generating factually incorrect or externally unsupported content [2].”
ImmiGo - An Adaptive Multilingual Chatbot for Immigration Assistance
Toriola, S., Chakraborty, A., & Frommholz, I. (2026) · Conference on Human Information Interaction and Retrieval · doi:10.1145/3786304.3787919
Bogers, T., Gäde, M., Hall, M. M., Koolen, M., Petras, V., & Skov, M. (2026) · Conference on Human Information Interaction and Retrieval · doi:10.1145/3786304.3787864 · Cited by 1
PRISM: Pyramidal representation and interaction synergy mechanism for malicious meme detection
Zhu, K., Yang, A., Li, Z., Zhang, Z., Wang, H., & Yu, Y. (2026) · Expert Systems with Applications · doi:10.1016/j.eswa.2026.131888
Application of large language models in medical interview training: a study with medical students
Hryszko, J., Michałek, A., & Roman, A. (2026) · BMC Medical Education · doi:10.1186/s12909-026-08819-2 · Cited by 2
“Hallucination and medical misinformation: LLMs are prone to generating plausible-sounding but factually incorrect information—a phenomenon known as “hal-lucination” [29].”
Large Language Models Readability Classification: A Variability Analysis of Sources and MetricsPreprint
de Matos, H. G. C., Wasmann, J.-W. A., Morata, T. C., Alvarenga, K. D. F., & JACOB, L. C. B. (2026) · medRxiv · doi:10.64898/2026.02.20.26346638 · Cited by 1
RegCheck-Hybrid: A Hybrid Rule-LLM Framework for Automated Multi-modal Fire Code Compliance Checking
Chen, Z., & Zhou, W. (2025) · International Conference on Mechatronic Engineering and Artificial Intelligence (MEAI) · doi:10.1109/meai68126.2025.11406521 · Cited by 1
“However, current Retrieval-Augmented Generation (RAG) technologies still face challenges in precise Numerical Reasoning, often leading to hallucinations when dealing with strict engineering constraints [4], [13].”
Consumer Profiling Leveraging Interval-Valued Clustering and LLM-Driven Prompt Generation
Li, Z., & Kabir, S. (2025) · International Conference on Algorithms, Computing and Artificial Intelligence (ACAI) · doi:10.1109/acai68217.2025.11406709
“For generating marketing texts and product recommendations, LLMs show great potential; however, their output often contains false information and lacks factual support [19], [20].”
A detailed review of large language models in the healthcare sector
Ozsahin, D. U., Emegano, D. I., Uzun, B., & Özsahin, İ. (2026) · Journal of Electrical Systems and Information Technology · doi:10.1186/s43067-026-00327-z
“It also advised that human feedback processes are incorporated into the LLMs [94].”
Yankouskaya, A., Almourad, M. B., Liebherr, M., Rahman, M. M., Alshakhsi, S. A. A., & Ali, R. (2026) · Research Square · doi:10.21203/rs.3.rs-9018418/v1 · Cited by 1
MACV: A Specialized Multi-Agent and Consensus Framework for Reliable LLM Outputs
More, R., Varma, S., & Varma, N. (2026) · International Conference on Mobile Computing and Sustainable Informatics (ICMCSI) · doi:10.1109/icmcsi67283.2026.11412804 · Cited by 1
“Comprehensive surveys identify intrinsic vs. extrinsic hallucinations, causes (decoding, dataset coverage, spurious correlations), and task-specific manifestations [7].”
Tuczek, M., Degirmenci, K., Breitner, M. H., Desouza, K. C., & Watson, R. T. (2026) · Business & Information Systems Engineering · doi:10.1007/s12599-026-00992-4 · Cited by 3
A Consistency-Oriented Verification Framework for Reliable Power Industry LLMs
Wang, B., Zhang, L., Li, Y., & Guo, N. (2025) · International Conference on Electronic Information Engineering and Computer Communication (EIECC) · doi:10.1109/eiecc67963.2025.11409479
“While these methods successfully reduce "knowledge blind spots"—enabling models to recognize industry-specific terminology—they do not inherently solve the problem of hallucinations [2].”
From essay to AI-augmented writing: reframing pedagogy and feedback through the AAWP model
Askari, M., & Rahim, A. S. (2026) · Assessment & Evaluation in Higher Education · doi:10.1080/02602938.2026.2638914 · Cited by 1
Yu, X. Y. X., Jiang, L., Zhuang, W., & Yang, L. (2026) · International Journal of Computational Intelligence and Applications · doi:10.1142/s1469026826500124
The Co-CreAItive Turn: Rethinking Teaching through AI Persona Frameworks
Andreadakis, Z. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-2637-5.ch003
ChatMicroscopy: A Perspective Review of Large Language Models for Next-Generation Optical Microscopy
Sancataldo, G. (2026) · Applied Sciences · doi:10.3390/app16052502
Knowledge Graph-Based Structural Safety Risk Diagnosis and Control of Prestressed Concrete Bridges
Hu, C., & Sun, Z. (2026) · Applied Sciences · doi:10.3390/app16052545
Translation, not Interpretation: Rethinking Language Model Design for Healthcare
Lamb, N. (2026) · SN Comprehensive Clinical Medicine · doi:10.1007/s42399-026-02316-9
Mohamed, A. R., Elmahjub, E., & Shaban, K. B. (2026) · Intelligent Systems with Applications · doi:10.1016/j.iswa.2026.200727 · Cited by 3
Large language models for urban transportation
Ma, Z., Wang, L., Qin, Z., & Ling, Y. (2026) · Elsevier eBooks · doi:10.1016/b978-0-443-26789-5.00011-0 · Cited by 2
Efficient-RAG: Competitive Single-Pass Performance in RAG via Strategic Context Reordering
Chen, Y., Huang, Z., & Zhang, Y. (2025) · International Conference on Advanced Computer Theory and Engineering · doi:10.1109/icacte66284.2025.11412212 · Cited by 1
“Despite their impressive capabilities, these models suffer from significant limitations like factual “hallucinations” [2] and static knowledge cutoffs [3].”
Evaluating the Performance of Large Language Models in Vascular Surgery Cases: Cross-Sectional Study
Wijetunga, A., Wang, Y., Yaghi, C., & Vicaretti, M. (2026) · JMIR Perioperative Medicine · doi:10.2196/91973
Understanding and Generating Student Questions with LLMs in Collaborative Learning
Kim, H. K., Nayak, S., Roknaldin, A., et al. (2025) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata66926.2025.11402653
“Despite ongoing concerns about hallucination in LLMs [40], [41], these findings suggest that the model appropriately leveraged the provided context in the prompt.”
Enhancing Static Analysis for ICC Resolution: An LLM-Integrated Approach
Jiang, Y., Gong, Q., & Gong, X. (2025) · International Conference on Advanced Computer Theory and Engineering · doi:10.1109/icacte66284.2025.11411955
“Here, stochasticity refers to the inherent unpredictability in model outputs [8], while hallucination manifests as LLMs generating factually ungrounded or meaningless content [9].”
Probing LLM Hallucination from Within: Perturbation-Driven Method via Internal Knowledge
Lee, S., Hsu, H., Chen, C., & Chau, D. H. (2025) · BigData Congress [Services Society] · doi:10.1109/bigdata66926.2025.11402288 · Cited by 1
“Detecting such hallucinations is crucial yet challenging due to their plausible appearance [4], [5] and diverse types [4], [6].”
Automated Structural Testing of LLM-Based Agents: Methods, Framework, and Case Studies
Kohl, J., Kruse, O., Mostafa, Y., Luckow, A., Schroer, K., Riedl, T. R., French, R., Katz, D., Luitz, M. P., Takher, T., et al. (2025) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata66926.2025.11401679 · Cited by 2
“[6] surveys hallucinations and defines it as "the generation of texts or responses that exhibit grammatical correctness, fluency, and authenticity, but deviate from the provided source inputs (faithfulness) or do not align with factual accuracy (factualness)".”
Sinha, N., Gupta, S., Kour, V., Misra, M., & Misra, R. N. (2025) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata66926.2025.11400870
“The phenomenon of “hallucination”, where LLMs generate factually incorrect or ungrounded information, is a significant and widely acknowledged challenge in the field [5].”
Thangaraj, S., Shukla, N. K., & Krishnamurthy, V. (2025) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata66926.2025.11402186
“Nevertheless, most end-to-end generative models suffer from hallucinations and lack factual grounding, which is problematic in safety-critical and knowledge intensive domains [7].”
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.




