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 3,101–3,200 of 4,844
Lazarev, A., Sedov, D., & Galkin, A. (2025) · International Conference on Control Systems, Mathematical Modeling, Automation and Energy Efficiency (SUMMA) · doi:10.1109/summa68668.2025.11302248
PRISM: Pyramidal Representation and Interaction Synergy Mechanism for Malicious Meme DetectionPreprint
Zhu, K., Li, Z. M., Yang, A., et al. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5985601
A Large Language Model Approach for In-Depth Qualitative Text Analysis
Astudillo, G. (2025) · International Conference on Pattern Recognition Systems (ICPRS) · doi:10.1109/icprs66293.2025.11302863 · Cited by 2
Prompt Chaining in Practice: A Case Study in Automated Scholarly Report Generation
Lazarev, A. (2025) · International Conference on Control Systems, Mathematical Modeling, Automation and Energy Efficiency (SUMMA) · doi:10.1109/summa68668.2025.11302303
“However, early approaches often rely on naive, single-shot prompting, which, when applied to complex synthesis tasks, can lead to superficial outputs, loss of critical context, and factual inaccuracies or "hallucinations" [3].”
Embedding Confidence to Enhance Trust in AI Document Entity Extraction
MacDonald, M., Khosravi, S., Ramin, A., & Meraji, S. (2025) · International Conference on Pattern Recognition Systems (ICPRS) · doi:10.1109/icprs66293.2025.11302825
“In high-stakes settings, such undetected errors can propagate bias and lead to unfair or even legally problematic decisions [4], [5].”
Dormitory Student Persona Development for A Generative AI Chatbot System
Tambunan, S. R., Simaremare, M. E. S., Simamora, A. N., Hutahaean, C., & Simamora, E. (2025) · International Conference on Informatics and Computing (ICIC) · doi:10.1109/icic68054.2025.11309523
“Implementation can be achieved through information augmentation approaches [7], integrating persona-specific conversational patterns and response structures into model training data to generate contextually appropriate responses aligned with each persona’s communication preferences and information…”
InfluentialHuman-LLM Collaboration Framework for Translating Health Instruments
Bitar, H., Almatrafi, O., Babour, A., Alzamzami, O., Alrige, M., & Alismail, S. (2025) · IEEE Access · doi:10.1109/access.2025.3649657 · Cited by 1
“Ji and colleagues [32] describe such behavior as a common limitation often arising from ungrounded pre-training data, noise, or over-reliance on language priors rather than factual grounding.”
Angelelli, L. (2025) · Brasiliensis · doi:10.64205/brasiliensis.14.27.2025.223
Вплив великих мовних моделей (LLM) на підходи до визначення поняття креативності
Тадеєв, Ю. (2025) · Педагогічні інновації ідеї реалії перспективи · doi:10.63437/3083-6433-2025-2(35)-17
Designing Prompts for Shared-Experience-Focused Utterance Generation in LLM-Based Dialogue Systems
Fukushima, R., & Hashimoto, M. (2025) · Journal of Machine Intelligence and Data Science · doi:10.11159/jmids.2025.009
Artificial Intelligence and Education: Potentials and Challenges
Filiz, O. (2025) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-8857-1.ch003
A Hybrid GraphRAG Framework for Geospatial Contextualization in Decision Support Systems
Germain, M., & Alayet, M. (2025) · IEEE/ACM International Conference on Utility and Cloud Computing (UCC) · doi:10.1145/3773274.3774662
Teo, T.-H., & Chow, Z.-J. (2025) · International Symposium on Embedded Multicore/Many-core Systems-on-Chip · doi:10.1109/mcsoc67473.2025.00026 · Cited by 1
“Accuracy and reliability remain primary concerns, with studies documenting instances of AI systems producing plausible-sounding but incorrect information, termed ”hallu-cinations” [19].”
Xie, H., Cao, H., & Zhao, H. (2025) · Systems · doi:10.3390/systems14010036
HypoSelectSimRAG-Enhancing Answer Accuracy in RAG via Multi-Path Self-Consistent Query Translation
Long, K., Huang, H., Zhao, X., & Tian, Z. (2025) · International Conference on Computer Engineering and Intelligent Control (ICCEIC) · doi:10.1109/icceic67916.2025.11309165
“Moreover, LLMs are prone to generating “hallucinations” [3]—content that appears plausible on the surface but is factually incorrect.”
AI Hallucination Prediction: A Novel Approach for Preventing False AI Outputs
Kundu, A., Malhotra, A., & Malhotra, V. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-06688-6_48
Neves, J. S., Pereira, M. C., & Bacalhau, L. M. (2026) · Studies in systems, decision and control · doi:10.1007/978-3-031-85398-2_44 · Cited by 1
Knowledge Augmented Finetuning Matters in Both RAG and Agent Based Dialog Systems
Wu, Y., Cai, Y., Huang, Y., Feng, J., & Ou, Z. (2026) · Communications in computer and information science · doi:10.1007/978-981-95-5382-2_20
Mitigating Risks in Large Language Model Usage Through Critical Thinking
Ziegfeld, L., Kox, E. S., Blok, J., van der Mijn, R., Venrooij, W., & van der Waa, J. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-13184-3_28
Gordon, N. A., Dixon, J., & Mian, Z. (2025) · Lecture notes in educational technology · doi:10.1007/978-981-95-5234-4_5
An Immersing Oriented Role-Playing Framework with Duplex Relationship Modeling
Wen, Y., Gao, S., Huang, C., Wang, Y., & Shang, S. (2026) · Lecture notes in computer science · doi:10.1007/978-981-95-4158-4_11
Shi, S., Ren, T., Lan, X., & Hu, J. (2026) · Lecture notes in computer science · doi:10.1007/978-981-95-3830-0_25
Nguyen, Q. H. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-00972-2_73
Efficient Tuning Framework for Resource- Constrained Biomedical Question Answering
Wang, B., Du, Y., Jin, X., & Wang, Z. (2026) · IEEE Transactions on Computational Biology and Bioinformatics · doi:10.1109/tcbbio.2025.3650547
“However, the training data and training strategies used in commercialized Large Language Models (LLM) are not public (e.g., ChatGPT), and the unpublic training data may contain false information that makes the models raise “hallucinations”[3], producing harmful results.”
Exploring the Impact of Large Language Models on Opinion Analysis in Arabic Social Media
Alzyadneh, M., Fraj, F. B., & Zribi, C. B. O. (2025) · IEEE/ACS International Conference on Computer Systems and Applications (AICCSA) · doi:10.1109/aiccsa66935.2025.11315188
“To mitigate this, techniques including ”reality grounding,” utilizing a variety of datasets, and adding verifiable data have been suggested as ways to guarantee factual accuracy and minimize inaccuracies [50].”
Generation of Synthesizable Verilog Code From Natural Language Specifications
Yashchenko, D. S., Romanov, A. Y., Ziazetdinov, A. A., Telpukhov, D. V., & Solovyev, R. A. (2026) · IEEE Access · doi:10.1109/access.2026.3651684 · Cited by 2
“The collected raw dataset was processed and cleaned using the following steps [34]: • all pairs were verified through automated compilation with Icarus Verilog; • modules requiring external libraries, IP cores, or test-benches were removed; • dataset closure was ensured, meaning that each pair can…”
Jin, L., Zhao, C., Han, D., Zuo, Z., Liang, Y., Liao, Y., & Tohti, T. (2026) · Knowledge and Information Systems · doi:10.1007/s10115-025-02676-z · Cited by 2
Araújo, P., Olympio, T., Soares, R. A., Barros, M. I., Silva, J., Ferreira, F. F. V. M., & Murari, T. B. (2025) · SAE technical papers on CD-ROM/SAE technical paper series · doi:10.4271/2025-36-0137
Saßmannshausen, T. M., & Wagener, S. (2026) · Qeios · doi:10.32388/gag6kd
IS TASK-BASED AI THE MISSING LINK BETWEEN HUMAN AND MACHINE REASONING?
Гончаров, С. С., Vityaev, E. E., Sviridenko, D., et al. (2025) · Journal of Mathematical Sciences · doi:10.1007/s10958-025-08144-x
Do LLMs Possess Strategic Intelligence? Testing LLMs in Iterated Prisoner’s DilemmasPreprint
Kommandeur, J., Vonk, M., Spiegeleire, S. D., & Sweijs, T. (2026) · Research Square · doi:10.21203/rs.3.rs-8405969/v1
Human- vs. AI-generated tests: dimensionality and information accuracy in latent trait evaluation
Angelelli, M., Oliva, M., Arima, S., & Ciavolino, E. (2026) · Statistics · doi:10.1080/02331888.2025.2610647 · Cited by 1
“In particular, the quality of AI outcomes can be severely affected by hallucinations [22], namely, generated data that are syntactically correct but express invalid, false, or partial information; without proper supervision and validation, they may seriously affect the quality of outputs when they…”
Pearson, S., Reyburn, M., Foley, C., et al. (2025) · doi:10.2196/preprints.90139
A Fine-Grained Model Editing Method Based on the Importance of Parameters
Wang, J., Song, C., & Ge, T. (2026) · IEEE Transactions on Audio Speech and Language Processing · doi:10.1109/taslpro.2026.3652007
“Nonetheless, LLMs have been criticized for frequently providing incorrect information [7], [8], [9], [10].”
Qamar, F., Latif, S., & Fatima, M. (2025) · International Conference on Frontiers of Information Technology (FIT) · doi:10.1109/fit67061.2025.11333596
“Its importance has grown alongside the rise of hallucinations, factual inconsistencies, and alignment failures in Large Language Models (LLMs) [1].”
Yang, K.-C., & Chu, H. (2025) · International Conference on Artificial Intelligence for Medicine, Health and Care (AIxMHC) · doi:10.1109/aixmhc65380.2025.00028
A Context-Aware Intelligent Tutoring Framework for Grammar Modeling Using Large Language Models
Daberao, L. (2025) · IEEE Conference on Pervasive and Intelligent Computing (PICom) · doi:10.1109/picom68402.2025.00034 · Cited by 3
“The issue of hallucinations in LLMs has been studied to a certain extent [16], [17].”
Du, R., Mancheong, F. U. N. G., Yuanjia, H. U., & Dongyang, L. I. U. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.01.12.699148
“However, this semantic flexibility comes at a cost: without structural grounding, LLMs are prone to "hallucination," generating plausible but factually incorrect associations due to the lack of explicit evidence 14,15 .”
Katwe, P., Balabantaray, R. C., & Vittala, K. P. (2026) · SN Computer Science · doi:10.1007/s42979-025-04684-w
Hidayati, N. N., & Atikah, L. (2025) · International Conference on Computer, Control, Informatics and its Applications · doi:10.1109/ic3ina68387.2025.11325645
“To address this, the Retrieval Augmented Generation (RAG) approach was developed as a primary architectural solution to mitigate such hallucinations in natural language generation [9], and its use has been systematically reviewed as an essential technique in high-stakes domains like healthcare [10].”
Influential周, 子. (2026) · Computer Science and Application · doi:10.12677/csa.2026.161010
Exploring LLM-Based Data Synthesis Strategies for Aligning Medical Consultation Preferences
Dou, C.-F., Zhang, Y., Jin, Z., Jiao, W., Zhao, H.-Y., Zhao, Y.-Q., & Tao, Z.-W. (2025) · Journal of Computer Science and Technology · doi:10.1007/s11390-025-4929-7 · Cited by 4
NIKI: Neighbor Interfering Knowledge Injection for Large Language Models
Ji, G., Jiang, Y., Wang, P., Yang, D., & Xia, Y. (2025) · IEEE International Conference on Parallel and Distributed Systems (ICPADS) · doi:10.1109/icpads67057.2025.11322877
“However, previous work [9] has also pointed out that when lacking domain-specific knowledge, LLMs are more prone to hallucinate in downstream tasks.”
Automatic Construction of Oilseed Rape Knowledge Graph Based on Large Language Models
Zhang, J., Zhang, H., Li, X., et al. (2025) · International Conference on Digital Society and Intelligent Systems (DSInS) · doi:10.1109/dsins68311.2025.11329855
“Prompt engineering techniques guide LLMs to extract structured knowledge from unstructured text without fine-tuning [5].”
LEAF: Logic-Enhanced Adaptive Fusion for Large-Scale Ontology Alignment
Zhu, X., Liu, B., Li, J., & Zhuang, X. (2025) · IEEE Intl Conf on Dependable, Autonomic and Secure Computing, Intl Conf on Pervasive Intelligence and Computing, Intl Conf on Cloud and Big Data Computing, Intl Conf on Cyber Science and Technology Congress (DASC/PiCom/CBDCom/CyberSciTech) · doi:10.1109/cyberscitech68397.2025.00058
“While demonstrating impressive reasoning capabilities, this end-to-end paradigm faces significant practical hurdles, including high API costs, inference latency, and a lack of control over the reasoning process, which can lead to factual “hallucinations” [21].”
Confidence Introspection: A Self-Reflection Method for Reliable and Helpful Large Language Models
Xi, T., Wang, C., & Zhang, J. (2026) · IEEE Transactions on Audio Speech and Language Processing · doi:10.1109/taslpro.2026.3654258 · Cited by 2
“LLMs may confidently generate responses while being unaware of their own errors [2], [36].”
The D3CNN text classification model based on multi-scale textual feature maps
Kang, C., & Li, K. (2025) · The Computer Journal · doi:10.1093/comjnl/bxaf151
Fernández, P. (2026) · Library Hi Tech News · doi:10.1108/lhtn-12-2025-0212
Multi-scale steering of large vision language models via visual information intervention
Zhao, D., Sun, B., He, J., & Zhang, Y. (2026) · Neurocomputing · doi:10.1016/j.neucom.2026.132780
Ethical issues with large language models
Tedesco, R. (2026) · Elsevier eBooks · doi:10.1016/b978-0-443-33592-1.00003-6
Constructing and Validating the AIPACK Scale: Measuring Teachers' AI Pedagogical Content Knowledge
Kuo, B., Wu, P., Fan, Y., & Liao, C.-H. (2026) · European Journal of Education · doi:10.1111/ejed.70464 · Cited by 2
Virtual Assistant System in Banking Application With Fine-Tuned LLM and RAG Approach
Gunawan, E., Amadeo, A. J., Kamardi, C., & Suhartono, D. (2025) · International Conference on Informatics, Multimedia, Cyber and Information System (ICIMCIS) · doi:10.1109/icimcis68501.2025.11327124 · Cited by 1
“Despite this advancement, language models still face several limitations, including inability to access up-to-date information [9], hallucination issues [10], [11], challenges with low-resource languages [12], and poor numerical reasoning abilities [13].”
RADAR: Agent-Guided Retrieval with Physics-Based Consistency Filtering for Aviation Fault Diagnosis
Liu, K., Zhang, S., Wang, Y., Zhang, H., & Zhang, W. (2025) · IEEE International Conference on Frontiers Technology of Information and Computer (ICFTIC) · doi:10.1109/icftic68075.2025.11324487 · Cited by 1
“Chain-of-thought prompting [7] elicits structured reasoning from LLMs, while Retrieval-Augmented Generation [8] grounds outputs in external knowledge bases to mitigate hallucinations [3].”
Applications of multi-agent systems in computational materials science
Ying, W., Zhong-Yi, L., & Zefeng, G. A. O. (2026) · Acta Physica Sinica · doi:10.7498/aps.75.20251687 · Cited by 1
DANV: A Dual-Agent Framework for Retrieval-Augmented News Verification
Chen, M. (2025) · International Conference on Cloud Computing, Big Data Application and Software Engineering (CBASE) · doi:10.1109/cbase67452.2025.11335506
“When confronted with dynamic news events that require external evidence for verification, they are prone to hallucinations [2], [3] and may deliver incorrect judgments.”
Integrating LLMs and Extended Reality for Interactive History-Telling Performances
Bochicchio, M. A., Miccoli, E., Prati, A., et al. (2025) · doi:10.1109/ieee-ch65308.2025.11279283
“…of Connecticut is exploring AI-generated scripts and their ethical implications through experimental theatrical research, while the Korean Design Research Society analyzes the impact of XR on identity and interactivity in performances, with applications in K-pop and fashion shows [12, 13].”
On the Reliability of Large Language Models in Financial Applications: An Analysis of Hallucination
Gupta, S. (2025) · International Conference on Applied Artificial Intelligence and Computing (ICAAIC) · doi:10.1109/icaaic64647.2025.11331105
“To address these challenges, a variety of hallucination mitigation strategies have been proposed in recent literature [2].”
A Comprehensive Analysis on Edge Deployment and Optimization of Llms
Liang, K. (2026) · Highlights in Science Engineering and Technology · doi:10.54097/jcb63r28
“Not all users have the expertise to verify the correctness of the outputs, misinformation spread to them can undermine the reliability of models’ input, and this can be the least severe consequences [10].”
Yin, W. (2026) · Creative Education · doi:10.4236/ce.2026.171007 · Cited by 2
Parycek, P., Sasdelli, D. C., & Edelmann, N. (2026) · Springer eBooks · doi:10.1007/978-3-658-49869-6_24
Mitigating Systematic Hallucinations in Large Language Models Using Dual-Process Cognitive Theory
Senturk, D., Bozanta, A., & Basar, A. (2025) · IEEE International Conference on Collaborative Advances in Software and Computing (CASCON) · doi:10.1109/cascon66301.2025.00117
“The growing use of Large Language Models (LLMs) in various applications faces a major obstacle: these models often ’hallucinate,’ producing responses that sound convincing but are actually incorrect or illogical [1], [2].”
ScenarioBench: Trace-Grounded Compliance Evaluation for Text-to-SQL and RAG
Atf, Z., & Lewis, P. R. (2025) · IEEE International Conference on Collaborative Advances in Software and Computing (CASCON) · doi:10.1109/cascon66301.2025.00111 · Cited by 1
“In parallel, RAG work has improved effectiveness and discussed hallucinations, yet provenance remains under-specified and unsupported attributions persist [2], [3], [5].”
CoScore: an LLM-Empowered Human-AI Collaborative Text Scoring Rule Generation and Calculation Engine
Song, Y. (2025) · IEEE International Conference on Teaching, Assessment, and Learning for Engineering · doi:10.1109/tale66047.2025.11346772
“However, directly using LLM for scoring poses risks such as uncontrollable hallucinations and unexplainable results [7].”
He, J., Huang, G., Yuan, X., Pun, C., Zhong, G., Yang, Q., Guo, L., Zhu, S., Lei, B., & Li, H. (2026) · Medical Image Analysis · doi:10.1016/j.media.2026.103958
Wang, X., Huang, Y., Liu, T., & Peng, S. (2026) · Lecture notes in electrical engineering · doi:10.1007/978-981-95-3320-6_17
Tiwari, A., & Gupta, V. (2026) · Research Square · doi:10.21203/rs.3.rs-8589677/v1 · Cited by 1
Peng, L., Xie, W., Gao, D., TIan, H., Yang, Z., Xian, N., Li, K., Liu, L., Zhao, Y., & Liu, Y. (2026) · doi:10.2196/preprints.91173
Zhou, Z., Gao, S., & Wang, Z. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6145233
IncrFuzz: LLM-Driven Incremental Fuzz Driver Generation for Library APIs
Feng, X., Gu, T.-T., Lu, S., & Kuang, X. (2025) · International Conference on Data Science in Cyberspace · doi:10.1109/dsc67331.2025.00093
“This strategy addresses a key challenge: LLMs often hallucinate [23]–[25] or misunderstand unfamiliar or complex APIs, resulting in harnesses that fail to compile [26].”
Lin, P., Li, J., Deng, H., Li, S., Yu, S., & Li, Y. (2025) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc58881.2025.11343559
“LLMs [32] often lack domain-specific knowledge [33], making RAG [34] an essential method to supplement LLMs with external information for high-quality responses.”
Gu, Y., Chen, X., Shan, G., Tao, J., Xia, Y., Gu, Y., Huang, P., & Shi, T. (2026) · BMC Emergency Medicine · doi:10.1186/s12873-026-01489-9
“Large language models (LLMs) have demonstrated substantial potential in optimizing clinical workflows and enhancing decision-making quality [4, 5].”
MedGuard-KG: Dual-Stream Knowledge Graph Verification for Safe Clinical Reasoning in Medical LLMs
Yitao, W., Kexin, M., Jiawen, D., & Fuji, R. (2025) · International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP) · doi:10.1109/iccwamtip68645.2025.11352596 · Cited by 1
“Two key issues are factual hallucinations and logical inconsistencies in generated content [3].”
Integrating AI Agents with Classical Reliability Assessment via LLMs of Semiconductor Lasers
Zhang, J., Wang, T., & Jiang, J. (2025) · International Conference on System Reliability and Safety Engineering (SRSE) · doi:10.1109/srse67406.2025.11357401
“At the same time, rigorous evaluations highlight persistent issues—such as hallucination, numerical instability, and overgeneralization—underscoring the need for constrained, verifiable outputs supported by retrieval and uncertainty checks [13], [14].”
Voth, V., Wade, J., Bierig, A., & Bertram, O. (2026) · AIAA SCITECH 2026 Forum · doi:10.2514/6.2026-0422
Yang, H., Jiang, J. Y., Huang, Q., Li, Y., & Tian, J. (2025) · International Conference on Energy Engineering and Power Systems (EEPS) · doi:10.1109/eeps68057.2025.11351591
Nawwar, M., Dewandaru, A., Saptawati, G. A. P., & Wicaksono, H. (2025) · IEEE International Conference on Data and Software Engineering (ICoDSE) · doi:10.1109/icodse68111.2025.11351857
“Recent advances in large language models (LLMs) enable semantic query understanding but remain vulnerable to hallucination when used without external grounding [8].”
Towards Robust Retrieval-Augmented Generation Based on Knowledge Graph: A Comparative Analysis
Amamou, H., Gagnon, S., Davoust, A., & Avila, A. R. (2025) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc58881.2025.11343466 · Cited by 1
“Nevertheless, these models still face serious problems such as factual hallucination [1], outdated knowledge [2], and absence of domain-specific expertise [3].”
Wang, N., Zhao, X., Lin, X., Liu, X., & Sun, Q. (2025) · IEEE International Conference on Bioinformatics and Biomedicine (BIBM) · doi:10.1109/bibm66473.2025.11356417
Guzzi, P. H., Carbonari, V., Canino, G., et al. (2025) · IEEE International Conference on Bioinformatics and Biomedicine (BIBM) · doi:10.1109/bibm66473.2025.11356771
“Moreover, LLMs suffer of the problem of hal-lucination , in which the model generates outputs that are fluent but factually incorrect [7], still limiting clinical application.”
Reason, Review, Repeat: Hybrid Chain of Thought to Mitigate Hallucinations in Large Language Models
Liu, J., Moh, M., & Moh, T.-S. (2026) · International Conference on Ubiquitous Information Management and Communication (IMCOM) · doi:10.1109/imcom69009.2026.11360964
“[6] explains that LLMs trained possibly have a bias related to optimization, while [5] emphasizes the mistake to tokenization or incorrect correlation due to similarity structure.”
Huang, X., Chen, Y., Feng, X., Zhu, H., & Tang, C. (2025) · International Conference on Power and Energy Applications (ICPEA) · doi:10.1109/icpea67476.2025.11349299 · Cited by 1
“…or unseen text and document collections in model training, it is often difficult for the LLMs alone to efficiently retrieve relevant information from a large amount of text, so much so that it generates outdated or factually inaccurate information, which is known as the "illusion" of the LLMs[2].”
Hu, W., Wang, M., Zhuang, X., Zhao, G., Zhang, P., & Zhang, G. (2025) · IEEE International Conference on Bioinformatics and Biomedicine (BIBM) · doi:10.1109/bibm66473.2025.11357159
“…the retrieved content, thereby undermining the reliability of the generated answers [10]; (2) When the retrieved evidence is incomplete or of low relevance, generative models are more likely to produce hallucinations or deliver answers that lack domain-specific accuracy and comprehensiveness [11].”
Computational Validation of AlphaFold 3 as a Design Engine for Next Generation Aptamer Therapeutics
Rayamajhi, M., Aryal, S., Gadhamshetty, V., et al. (2025) · IEEE International Conference on Bioinformatics and Biomedicine (BIBM) · doi:10.1109/bibm66473.2025.11356635
“…benchmark, it is impossible to distinguish whether AF3 is generating accurate physical models or merely plausible structural ”hallucinations” [2] that will fail in these complex biological milieus, a concern that has also been observed in recent ML-based protein prediction studies in…”
Mahapatra, J., & Garain, U. (2026) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3795137 · Cited by 2
Design and Prototype of a Chatbot for Public Participation in Major Infrastructure Projects
Matthei, J., Maas, J., Wischum, M., Mackenbach, S., & Klemt‐Albert, K. (2026) · Multimodal Technologies and Interaction · doi:10.3390/mti10020012
Transparent and Equitable Metrics and Models
Arias-Duart, A., Cortés, Á., & Cortés, U. (2026) · Springer eBooks · doi:10.1007/978-3-031-61050-9_28-1
An LLM-guided structural optimization method for aero-engine lubrication hydrocyclones
Shi, C., Yang, D., Jianping, L., Zhao, M., Yi, Y., Wang, H., & Jiang, X. (2026) · Separation and Purification Technology · doi:10.1016/j.seppur.2026.137096 · Cited by 2
Li, A., Bi, X., Chen, S., Hu, J., & Shi, Y. (2026) · Health Information Science and Systems · doi:10.1007/s13755-026-00432-3
“Furthermore, persistent issues—includ-ing hallucinations arising from knowledge gaps and deficiencies in natural language generation [34], as well as catastrophic forgetting caused by overparameterization and high task similarity—continue to constrain LLM reliability.”
JADE: Jawbone Lesion Diagnosis and Decision Supporting SystemPreprint
Saadi, S. B., Berne, J. V., Fontenele, R. C., et al. (2026) · medRxiv · doi:10.64898/2026.01.26.26344704
Wright, B. S., Kim, L. J., & Coleman, N. R. (2023) · Annals of Pharmacy Education Safety and Public Health Advocacy · doi:10.51847/hpn7ecjdva
Industrial-Scale Data Processing and LLM Evaluation for Coal Mine Safety
Yang, M., Cui, C., Li, Y., Chen, N., & Zhang, P. (2025) · International Conference on Communication and Information Processing (ICCIP) · doi:10.1145/3784833.3784879 · Cited by 1
Defending LLMs against jailbreak attacks through representation offset detection
Liu, S., Cheng, X., Zheng, Z., & Su, S. (2026) · Information Processing & Management · doi:10.1016/j.ipm.2026.104662 · Cited by 2
Kauro, a graph-based chatbot for high-fidelity information transmission conversationsPreprint
King, C. H., Barrick, R., Almalvez, M., Blanco, K., Dios, I. D., Fusaro, V. A., Délot, E. C., Donohue, C., Berger, S. I., Xiao, C., et al. (2026) · medRxiv · doi:10.64898/2026.01.30.25342358 · Cited by 1
“Such guardrails could allow participants to ask clarifying questions in their own words, while ensuring that answers are drawn from pre-approved sources and never modify the core consent script 38,41 .”
García-Campos, J. M., Lara-Romero, A. W., Mayor, V., & Calvillo‐Arbizu, J. (2026) · Data · doi:10.3390/data11020029 · Cited by 2
“Furthermore, LLMs may rely on outdated or incomplete sources [5], potentially leading to the dissemination of incorrect knowledge.”
Knowledge Reranking Augmented LLM for Industry Accident AnalysisPreprint
Cao, Q., Ning, J., Ma, J., & Zhang, C. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6170128
Fact-Checking with Large Language Models via Cost-Effective First-Order Logic Reformulation
Asghari, S., Lakshmanan, L. V. S., Srinivasan, V., & Thomo, A. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-14107-1_6
From Keywords to Context: An AI Agent for Natural Language Document Lookup in the Enterprise
Ngoc, H. L. (2025) · RIVF International Conference on Computing and Communication Technologies (RIVF) · doi:10.1109/rivf68649.2025.11365175
“LLMs trained on general web data lack domain-specific knowledge and are prone to "hallucinations" — generating plausible but factually incorrect or un-verifiable information [3], [4].”
DoraReviewer: Vietnamese Virtual Reviewer with GraphRAG and LLM
Trinh, T. T. N., Ngo, A. H. L., Nguyen, T. H. T., & Ly, K. D. D. (2025) · RIVF International Conference on Computing and Communication Technologies (RIVF) · doi:10.1109/rivf68649.2025.11365128
“Existing generation models often lack grounding in reliable data, leading to factual inconsistencies and hallucinations that can mislead users [9, 10].”
Trang, T., Tran, P., Le, T. M., Le, L., Maheshwari, G., Do, N., & Dao, S. V. T. (2025) · RIVF International Conference on Computing and Communication Technologies (RIVF) · doi:10.1109/rivf68649.2025.11365126
“Limitations include hallucination [12], [13], outdated knowledge [14], and limited transparency or personalization.”
Multi-Perspective Visual Contrastive Decoding for Reliable Assistance
Pan, B., Shi, H., & Gao, X. (2026) · ACM Transactions on Internet of Things · doi:10.1145/3785360
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.




