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 2,201–2,300 of 4,844
Yu, M., Wang, J., Chen, Y., Lepech, M. D., Liu, Y., & Law, K. H. (2025) · IEEE Annual International Computer Software and Applications Conference (COMPSAC) · doi:10.1109/compsac65507.2025.00081 · Cited by 1
“, [1, 0, 1], [2]) Make sure the answer is complete in no more than 150 words.”
Yang, Y., Li, C., Han, Z., Li, R., Xu, K., Li, Q., Zhong, W., Shen, Z., Fei, Z., Ge, J., et al. (2026) · ACM Transactions on Software Engineering and Methodology · doi:10.1145/3764584 · Cited by 12
Hwang, K., Kim, M., Kim, M. A., Park, C., Park, Y., Lee, C., Lim, J., & Oh, H. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.129548 · Cited by 1
Yadav, O., & Andreopoulos, W. (2025) · IEEE International Conference on Big Data Computing Service and Applications (BigDataService) · doi:10.1109/bigdataservice65758.2025.00036 · Cited by 1
“Ji et al. [3] extensively document hallucinations—plausible-sounding but factually incorrect outputs generated by LLMs.”
Reinforcement Learning for Constraint-Aware Knowledge Base Question Answering
Yang, J., & Yu, K. (2025) · IEEE Access · doi:10.1109/access.2025.3606830 · Cited by 2
“However, despite their impressive capabilities, LMs are plagued by notable challenges, including issues such as hallucinations [1], [2] and weaknesses in numerical reasoning [3], [4], [5].”
The 13th International Workshop on News Recommendation and Analytics (INRA 2025)
Iana, A., Treuillier, C., Yadav, V., Kille, B., Lommatzsch, A., & Özgöbek, Ö. (2025) · ACM Conference on Recommender Systems (RecSys) · doi:10.1145/3705328.3747997 · Cited by 1
The Case for Using Generative AI to Run Deliberation Simulations
Rountree, J., & Gastil, J. (2026) · Journal of Deliberative Democracy · doi:10.16997/jdd.1625 · Cited by 6
Multi-agent Retrieval-Augmented Generation for Enhancing Answer Generation and Knowledge Retrieval
Kumar, D., & Jain, B. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-05179-0_22 · Cited by 5
Parse Trees Guided LLM Prompt Compression
Mao, W., Hou, C., Zhang, T., Lin, X., Tang, K., & Lv, H. (2025) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2025.3609956 · Cited by 10
“These generative compression methods exploit the language understanding and generation ability of language models, but would encounter issues like hallucination [11] due to the limitations of generative language models.”
Comparative Evaluation of Four Large Language Models in Turkish Dentistry Specialization Exam
Ekici, Ö. (2025) · Selcuk Dental Journal · doi:10.15311/selcukdentj.1674113 · Cited by 2
Dilbaz, O. F., Ozates, M. N., Bolat, B., Gündüz-Demir, Ç., & Kulaç, İ. (2025) · American Journal of Clinical Pathology · doi:10.1093/ajcp/aqaf091 · Cited by 2
“25,26 To alleviate this issue, we provided clear prompts, giving domain-specific instructions and options for each parameter to conform outputs to predefined templates.”
Multi-agent Reasoning for Cardiovascular Imaging Phenotype Analysis
Zhang, W., Qiao, M., Zang, C., Niederer, S., Matthews, P. M., Bai, W., & Kainz, B. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-04927-8_41 · Cited by 12
Generative AI Efficiency and Effectiveness in Software Project Documentation Review Process
Demir, K. A., Muppalla, T. N. V., & Liu, B. (2025) · International Conference on Artificial Intelligence, Computer, Data Sciences and Applications (ACDSA) · doi:10.1109/acdsa65407.2025.11165833 · Cited by 1
Patil, S. S., Ramteke, M., Verma, M., Chandra, T., & Rathore, A. S. (2025) · Neural Computing and Applications · doi:10.1007/s00521-025-11600-z · Cited by 2
Empty Universities: The Dystopian Future of Education by AI
Sparrow, R., & Flenady, G. (2025) · Springer eBooks · doi:10.1007/978-981-96-7185-4_6 · Cited by 1
Health-oriented Multimodal Food Question Answering with Implicit and Explicit Knowledge
Hu, M., Song, Y., Yang, X., Wang, Y., & Xu, C. (2025) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3766065 · Cited by 1
“Implicit knowledge, sourced from pretrained models like GLM-130B [68] and GPT-4 1 , offers extensive general knowledge, though these models may struggle with factual accuracy and occasionally generate knowledge at odds with reality [22].”
IILAP: Interactive Information Literacy Assessment Platform
Dadic, P., & Магдановна, Е. Л. (2025) · Communications in computer and information science · doi:10.1007/978-3-032-06136-2_28 · Cited by 1
Leveraging Large Language Models for Smart Educational Services: A Comprehensive Review
Li, Y., Liu, T., & Yang, W. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-06524-7_7 · Cited by 3
Chang, Q., Li, C., Bhatta, K., & Waseem, M. (2025) · Journal of Computing and Information Science in Engineering · doi:10.1115/1.4070032 · Cited by 4
ControlA: Agentic Workflow Control Mechanisms for Reliable Science
Gueroudji, A., Mallick, T., de Souza, R. P., da Silva, R. F., Ross, R. B., Dorier, M., Carns, P., Chard, K., & Foster, I. T. (2025) · IEEE International Conference on eScience · doi:10.1109/escience65000.2025.00086 · Cited by 6
“Moreover, these agents are prone to generating hallucinated or mis-leading content [15].”
Human-Centered Business Education in an Artificial Intelligence-Driven World
Fukami, C. V., Hamilton, A. L., Rivers, C., Holland, A., Brady, M., Fellenz, M. R., Allen, S. J., Caporarello, L., Fischbacher‐Smith, D., Fischbacher‐Smith, M., et al. (2025) · Journal of Management Inquiry · doi:10.1177/10564926251364213 · Cited by 2
Fostering AI alignment through blockchain, proof of personhood and zero knowledge proofs
Neulinger, A., & Sparer, L. (2025) · Cluster Computing · doi:10.1007/s10586-025-05729-8 · Cited by 5
“Classifying, filtering and limiting the noise in ML model input data by correctly annotating it reduces the AI safety risks [ 101].”
InfluentialChen, Z., Li, A., Zhang, N., Chen, J., Huang, Y., & Zheng, Z. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.129972 · Cited by 4
Trailblazing specific generative models (SGMs) for early-stage engineering design concepts
Obieke, C. C., Milisavljevic-Syed, J., Jiang, P., Bridgeman, J. O., & Han, J. (2025) · Design and Artificial Intelligence · doi:10.1016/j.daai.2025.100030 · Cited by 6
Using AI to assess the decision-usefulness of corporates’ nature-related disclosures
Senni, C. C., Vaghefi, S. A., Manekar, T., Schimanski, T., & Leippold, M. (2025) · Environmental Research Communications · doi:10.1088/2515-7620/ae1357 · Cited by 9
Foundation Model for Medical Imaging: A Comprehensive Review
Jiao, L., Yang, J., Li, R., Liu, F., Liu, X., Chen, P., Guo, Y., Li, L., Shang, R., & Ma, W. (2025) · IEEE Transactions on Artificial Intelligence · doi:10.1109/tai.2025.3618796 · Cited by 11
“These real-world challenges lead to a fragile MFMs and the propensity to generate unexpected hallucinations when facing new tasks [26], making it difficult to realize the vision of GMAI.”
Shang, R., Mallari, K., Yeong, W. B. A., Yasuhara, K., Tang, A., & Hsieh, G. (2025) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3757501 · Cited by 3
Hermes: Efficient Serving of LLM Applications with Probabilistic Demand Modeling
Liu, Y., Gan, Z., Gan, Z., Wang, W., Chen, C., Shan, Y., Chen, X., Han, Z., Zhu, Y., Sun, S., et al. (2026) · ACM Transactions on Architecture and Code Optimization · doi:10.1145/3803390 · Cited by 7
“Meanwhile, the output of an LLM request may be unreliable (i.e., suffering hallucination [42]), and additional LLM requests would be required to ensure ∗ Chen Chen is the corresponding author. output quality (e.g., with self-reflection [43]).”
Shi, C., & Paracha, S. (2025) · Mental Health and Digital Technologies · doi:10.1108/mhdt-07-2025-0041 · Cited by 2
Generating Robotic Control Strategies With LLMs : Via Human–Robot Voice Interaction
Wang, X., Zhou, G., Hua, H., Tang, Z., & Cao, Y. (2025) · Concurrency and Computation Practice and Experience · doi:10.1002/cpe.70372 · Cited by 2
SHALE: A Scalable Benchmark for Fine-grained Hallucination Evaluation in LVLMs
Yan, B., Chen, Z., Min, Y., Zhang, J., Wang, J., Wang, X., & Shan, S. (2025) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3746027.3758308 · Cited by 4
Large Language Models for chatbot applications handling sensitive information
Oliveira, H., Sobrinho, Á., Dias, A. D. R. C., de Araújo, A. W. M., Araújo, R. D., Matos, D. D. M. D. C., & Galvez, S. M.-N. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.130145 · Cited by 6
Enhancing Brazilian Legal Information Retrieval: An Automated Keyphrase Generation
Funicheli, B. O., Sakiyama, K. M., Nogueira, R. F., & Romero, R. A. F. (2025) · Journal of the Brazilian Computer Society · doi:10.5753/jbcs.2025.5711 · Cited by 1
AI and Agile Software Development: From Frustration to Success XP2025 Workshop Summary
Herda, T., Pichler, V., Zhang, Z., Abrahamsson, P., & Hanssen, G. K. (2025) · Lecture notes in business information processing · doi:10.1007/978-3-032-05799-0_1 · Cited by 8
AdaGCRAG: Adaptive Graph-Chunk Retrieval for Lightweight RAG
Zhang, Y., Xu, Z., Huo, D., Guo, X., Zhou, Q., & Zhang, Y. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-09527-5_32 · Cited by 2
Multi-Level Graph Pruning-Based Framework for Graph Retrieval-Augmented Generation
Li, H., Li, X., Su, F., & Guo, Q. (2025) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme59968.2025.11209364 · Cited by 2
“Consequently, this dependency can result in inconsistencies and factual inaccuracies, particularly in open-ended question answering and automated summarization tasks [5].”
A Design of Experiments Oracle LLM for Research-Grounded Combinatorial Testing
Margabandu, L., Chen, Z., Wong, W. E., & Hsu, C.-W. (2025) · IEEE International Conference on Software Quality, Reliability and Security Companion (QRS-C) · doi:10.1109/qrs-c65679.2025.00021 · Cited by 1
“When presented with questions outside the domain, the Oracle appropriately issued clarification responses or refrained from answering—an expected behavior in retrieval-augmented frameworks designed to reduce unsupported generations [21].”
Qing, S., & Chen, L. (2025) · International Conference on Educational Knowledge and Informatization (EKI) · doi:10.1145/3765325.3765345 · Cited by 1
Ding, Y., Lyu, H., Cao, Y., Song, W., Tian, L., & Liu, S. (2025) · Chinese Journal of Electronics · doi:10.23919/cje.2025.00.042 · Cited by 4
M2RAG: A Multi-agent and Multimodal Fusion Framework for Retrieval-Augmented Document QA
Duan, T., Hu, M., Xue, L., Liu, C., Geng, G., Luo, W., & Luo, Z. (2025) · Communications in computer and information science · doi:10.1007/978-981-95-4088-4_35 · Cited by 2
Homophily-induced emergence of biased structures in LLM-based multi-agent AI systems
Mehdizadeh, A., & Hilbert, M. (2025) · Social Network Analysis and Mining · doi:10.1007/s13278-025-01535-7 · Cited by 17
“These hallucinations indicate a failure in the model’s ability to strictly adhere to task constraints, which is critical in controlled node selection settings [127, 128].”
Leveraging the Graph-Based LLM to Support the Analysis of Supply Chain Information
Su, P., Xu, R., & Chen, D. (2025) · Informatics · doi:10.3390/informatics12040124 · Cited by 3
“In particular, hallucination refers to a type of response that is coherent and grammatically correct but factually or semantically incorrect [10].”
Large language model-enhanced probabilistic modeling for effective static analysis alarms
Pan, X., Li, J., Zhou, Z. H., Li, G., Chen, X., Ma, J., Wu, J. F., & Zhang, Q. (2025) · Frontiers of Information Technology & Electronic Engineering · doi:10.1631/fitee.2500038 · Cited by 4
AI in Conflict Resolution: Practical Considerations, Opportunities and Challenges
Molak, A. (2025) · Conflict Resolution Quarterly · doi:10.1002/crq.70012 · Cited by 1
TravQuery: A Customer Support Chatbot Based on Retrieval Augmented Generation (RAG)
Farnaz, A., & Huyck, C. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-11402-0_10 · Cited by 1
Jin, Z., Yi, Z., Tu, Z., & Chu, D. (2025) · Service Oriented Computing and Applications · doi:10.1007/s11761-025-00481-8 · Cited by 1
“While KGP demonstrates the potential of LLMs in capturing complex relationships within the graph, it also faces significant challenges, such as the risk of hallucination [12] and limited depth in reasoning, which can lead to incomplete or inaccurate answers.”
Social and Ethical Risks Posed by General-Purpose LLMs for Settling Newcomers in Canada
Nejadgholi, I., Molamohammadi, M., & Bakhtawar, S. (2025) · Women in engineering and science · doi:10.1007/978-3-031-95890-8_11 · Cited by 5
LLMs in Staging: An Orchestrated LLM Workflow for Structured Augmentation with Fact Scoring
Trimigno, G., Lombardo, G., Tomaiuolo, M., Cagnoni, S., & Poggi, A. (2025) · Future Internet · doi:10.3390/fi17120535 · Cited by 4
da Cunha, M. V., Silveira, M. R., Santana, B. S., de Freitas, L. A., & Corrêa, U. B. (2025) · Brazilian Symposium on Multimedia and the Web · doi:10.5753/webmedia.2025.16029 · Cited by 4
Lightweight Self-Consistency Voting Framework for Reliable Hallucination Detection in LLMs
Padmavathi, A., Pandit, B., & Khaitan, G. (2025) · International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO) · doi:10.1109/icrito66076.2025.11241896 · Cited by 1
“…from a specialized research question to a significant bottleneck for reliable NLP. Early mapping surveys like Huang et al. [12] and Ji et al. [13] methodically charted the topography, dividing hallucinations into intrinsic and extrinsic, and noting such types as factual hallucinations,…”
Teng, J., & Cho, S. (2026) · Applied Research in Quality of Life · doi:10.1007/s11482-026-10606-6 · Cited by 4
Peng, Z., Liu, X., Yang, G., Liu, J., Peng, X., & Long, Y. (2025) · Pattern Recognition Letters · doi:10.1016/j.patrec.2025.11.040 · Cited by 2
Wang, C., Cao, Z. G., & Zhou, K. (2025) · ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD) · doi:10.1109/snpd65828.2025.11253421 · Cited by 2
Learnable Game-Theoretic Policy Optimization for Data-Centric Self-Explanation Rationalization
Zhao, Y., Wang, Z., Yu, X., Li, X., Liang, J., & Li, R. (2025) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2025.3638864 · Cited by 10
“However, due to unpredictable failure problems [47] and hallucinated reasoning [48], CoT-based generative explanations produced by LLMs are often unreliable and not always faithful in high-stakes scenarios [49], [50].”
LLMQR: efficient knowledge distillation for questionnaire recommendation via large language models
Huang, Z., Long, R., Chen, H., Wu, M., Li, Q., & Na, J. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.130632 · Cited by 3
Berro, C., Deligiannaki, F., Stefani, T., Christensen, J. M., Gerdes, I., Köster, F., Hallerbach, S., & Raulf, A. (2025) · IEEE/AIAA Digital Avionics Systems Conference (DASC) · doi:10.1109/dasc66011.2025.11257297 · Cited by 2
“Since hallucinations are inevitable once LLMs are used as general problem solvers [39], [40], it is very important to note that solutions are neither created nor modified by the LLM.”
MINDS: An NLP-Driven Process Mining Framework for Superior Microservices Decomposition
Nassima, A. M., Sbaï, H., & Bäına, K. (2025) · IEEE Access · doi:10.1109/access.2025.3639460 · Cited by 3
“Potential safeguards include: (1) Prompt engineering with explicit constraints and validation rules [83], [84]; (2) Multi-prompt verification with consistency checking; (3) Retrieval-Augmented Generation (RAG) approaches grounding responses in verified domain knowledge; (4) Human-in-the-loop…”
The Application of Foundation Models in Radiology: Bridging Images, Reports, and Beyond
Lee, D. (2025) · Journal of the Korean Society of Radiology · doi:10.3348/jksr.2025.0048 · Cited by 2
“When a foundation model generates radiology reports or diagnostic conclusions, it may produce “hallucinations”—instances where the model describes findings that do not exist or issues diagnoses for conditions not supported by the input data[58].”
A Framework of Knowledge Graph-Enhanced Large Language Model Based on Global Planning
Li, Y., Song, D., Tian, Y., Wang, H., Zhou, C., & Zhang, S. (2025) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2025.3639599 · Cited by 8
“A LTHOUGH large language models (LLMs) possess out-standing language understanding and generation capabilities [1], they still suffer from issues such as lack of inter-pretability and hallucination [2].”
Arthur, A. W., Moore, A., Carmona-Rosado, L., Kassam, P., Logel, M., Nieves, C. I., Tope, P., Litaker, D. G., & Borrell, L. N. (2025) · Preventive Medicine Reports · doi:10.1016/j.pmedr.2025.103334 · Cited by 2
“Technologies such as LLMs are known to generate “ authoritative-sounding content, ” which may lack a holistic evidentiary view, and can result in outputs of questionable accuracy (Elsevier, 2025b) and outright falsifications (Cheng et al., 2025; Emsley, 2023; Ji et al., 2023; Mitchell, 2025).”
Predictive hermeneutics: bias, culture, and the predictive mind
Devereaux, J. J. (2025) · Phenomenology and the Cognitive Sciences · doi:10.1007/s11097-025-10126-y · Cited by 1
Automatic Minds: Cognitive Parallels Between Hypnotic States and Large Language Model Processing
Riva, G., Wiederhold, B. K., & Mantovani, F. (2025) · Cyberpsychology Behavior and Social Networking · doi:10.1177/21522715251400733 · Cited by 2
“17 Hallucinations arise because the model ’ s automatic pattern-completion mechanism cannot distinguish between statistically likely and factually accurate continuations.”
Blinded evaluation of GPT-4 and physician responses to patient inquiries across multiple specialties
Sallabi, A., Hrvoj, H., & Štajduhar, A. (2025) · Digital Health · doi:10.1177/20552076251390001 · Cited by 3
Language models for the analysis of and interaction with climate change documents
Volkanovska, E. (2025) · Environmental Data Science · doi:10.1017/eds.2025.10024 · Cited by 3
“84 A type of error identified in the output of generative LMs, which refers to generated content that is unsensical or unfaithful to the provided source content (Ji et al., 2023).”
Strategic Framework for Evaluating Curriculum-Job Fit via Knowledge-Injected Large Language Models
Lee, H., & Yi, S. (2025) · IEEE Access · doi:10.1109/access.2025.3643626 · Cited by 1
“This indicates that KI improved the model’s ability to distinguish course-level competencies within and across majors, consistent with LLM evaluation literature emphasizing contextual precision and reduced semantic drift [26], [27].”
Snip-Cache: A code snippet caching system for LLM-based command-driven IoT systems
Song, C., & Kang, S. (2025) · Internet of Things · doi:10.1016/j.iot.2025.101852 · Cited by 2
A Hybrid Neuro-Symbolic Reasoning Framework for Automated Factual Verification in Conversational AI
Chin, W. H., Guo, Y., Loo, C. K., & Kubota, N. (2025) · International Conference on Machine Learning and Cybernetics · doi:10.1109/icmlc66258.2025.11280070 · Cited by 18
Are you hallucinated? Insights into large language models
Singh, S., Saha, R., Kumar, G. K., Nayyar, A., & Kim, T.-K. (2025) · ICT Express · doi:10.1016/j.icte.2025.12.011 · Cited by 3
Mezzadri, M., & Paita, M. (2025) · Italiano LinguaDue · doi:10.54103/2037-3597/30456 · Cited by 2
New Mind Ai Team (2025) · Procedia Computer Science · doi:10.1016/j.procs.2025.08.239 · Cited by 4
Yao, S., Wu, S., Yuan, C., et al. (2025) · Materials Today Communications · doi:10.1016/j.mtcomm.2025.113915 · Cited by 1
Wang, R., Yang, L., Yang, Q., & Cao, C. (2025) · Agronomy · doi:10.3390/agronomy15081958 · Cited by 1
Daten, Datenmanagement und Datenauswertung
Leimeister, J. M. (2025) · Springer eBooks · doi:10.1007/978-3-662-70646-6_3 · Cited by 1
Kou, J., Lu, X., Yin, A., Zhang, W., Fang, X., Li, Y., Pan, T., Zhang, L., Cheng, G., Yuan, H., et al. (2024) · International Conference on Universal Village (UV) · doi:10.1109/uv63228.2024.11189154 · Cited by 7
GADE+: A Graph-Based Anchor-Enhanced Framework for Targeted Document Detection
Wen, H., Shen, W., & Li, G. (2025) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2025.3621835 · Cited by 1
“Additionally, to address the hallucination problem [6] of large language models (e.g., GPT-3 [7] and PaLM [8]), a series of recentresearchendeavorsnamedretrievalaugmentedgeneration (RAG)[9],[10],[11],[12],[13]haveattemptedtoretrieve externalWebdocumentstoaugmentlargelanguagemodelsand…”
Not a Swiss Army Knife: Academics' Perceptions of Trade‐Offs around Generative AI Use
Razi, A., Bouzoubaa, L. A., Pessianzadeh, A., Seberger, J. S., & Rezapour, R. (2025) · Proceedings of the Association for Information Science and Technology · doi:10.1002/pra2.1277 · Cited by 5
Automata-Based Steering of Large Language Models for Diverse Structured Generation
Luan, X., Wei, Z., Zhang, Y., & Sun, M. (2025) · Lecture notes in computer science · doi:10.1007/978-981-95-4213-0_2 · Cited by 4
Dynamic and Parametric Retrieval Augmented Generation
Su, W., Dong, Q., Ai, Q., & Liu, Y. (2025) · Annual International ACM SIGIR Conference on Research and Development in Information Retrieval in the Asia Pacific Region (SIGIR-AP) · doi:10.1145/3767695.3769672 · Cited by 12
Challenges for Computational Reliabilism in AI and Other Computational Methods
Alvarado, R. (2025) · Synthese Library/Synthese library · doi:10.1007/978-3-032-03083-2_5 · Cited by 4
RepreGuard: Detecting LLM-Generated Text by Revealing Hidden Representation Patterns
Chen, X., Wu, J., Yang, S., Zhan, R., Wu, Z., Luo, Z., Wang, D., Yang, M., Chao, L. S., & Wong, D. F. (2025) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl.a.61 · Cited by 30
Narayan, A., Kumar, S. D. M., & Chacko, A. M. (2025) · Telematics and Informatics Reports · doi:10.1016/j.teler.2025.100285 · Cited by 8
Echo-LLM Evidence-Checked Hierarchical Ontology
Dalal, A. S., & McGinty, H. (2025) · Open Conference Proceedings · doi:10.52825/ocp.v8i.3173 · Cited by 1
“Yet their propensity for factual hallucination [8]—producing fluent but false assertions—renders them unreliable for scientific and enterprise applications [9].”
Makino, M., She, W. J., Siriaraya, P., Matoba, S., & Senoo, K. (2025) · Scientific Reports · doi:10.1038/s41598-025-34079-z · Cited by 1
Large AI Models Have a Prioritization Problem: Policy Implications and Solutions
Jackson, J. C., Liu, Y., Wang, Z., & Brady, W. J. (2025) · Policy Insights from the Behavioral and Brain Sciences · doi:10.1177/23727322251408311 · Cited by 2
“In LLMs, emphasizing coherence results in hallucinations — fl uent but factually inaccurate statements (Ji et al., 2023).”
Awad, A., & Alahmari, D. A. (2025) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-7006-4.ch006 · Cited by 1
Can Large Language Models Assist with SOTIF Scenario Generation?
Cyffka, E., Diemert, S., Adiththan, A., Debouk, R., Ramesh, S., Kernot, J., & Joyce, J. J. (2025) · IEEE International Conference on Recent Advances in Systems Science and Engineering (RASSE) · doi:10.1109/rasse64831.2025.11315433 · Cited by 1
“LLMs occasionally exhibit a phenomenon referred to as “hallucination”, in which they produce outputs that appear to be plausible, but are, in fact, inaccurate [5].”
LeCov: Multi-level testing criteria for large language models
Xie, X., Song, J., Huang, Y., Song, D., Juefei-Xu, F., & Ma, L. (2025) · Journal of Systems and Software · doi:10.1016/j.jss.2025.112763 · Cited by 3
Jang, D.-S., Yi, J.-S., Jeon, H., & Hong, Y. (2025) · Sustainability · doi:10.3390/su18010396 · Cited by 3
Smarter Coaching Chatbots: Leveraging Hybrid Architectures for Deeper Context Awareness
Bauer, A., Mai, V., Neef, C., & Richert, A. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-13184-3_1 · Cited by 2
Do Large Language Models Know When They Lack Knowledge?
Qin, S., Zhou, L., Sun, L., & wang, N. (2026) · Electronics · doi:10.3390/electronics15020253 · Cited by 3
Accurate Table Question Answering with Accessible LLMs
Jiang, Y., Wei, F., Bao, E., Li, Y., Ding, B., Yang, Y., & Xiao, X. (2026) · IEEE International Conference on Data Engineering (ICDE) · doi:10.1109/icde65706.2026.00020 · Cited by 3
“While the logic and query agents follow distinct reasoning paths, they must interact strategically to maintain consistency and avoid error propagation, e.g., due to hallucinations [27].”
Envisioning generative artificial intelligence in cartography: mapmaking, map use, and ethics
Kang, Y., & Wang, C. (2026) · International Journal of Cartography · doi:10.1080/23729333.2025.2582231 · Cited by 8
“Closely related to confabulation is hallucination , which refers to the generation of content that does not exist in the model’s input data or contradicts known facts (Ji et al., 2023).”
GÜLDEN, B., Bilge, H., & Uysal, P. K. (2026) · International Journal of Assessment Tools in Education · doi:10.21449/ijate.1678002 · Cited by 4
“Such recycling of templates is consistent with what has been described in the literature as template-based generation in large language models, where outputs rely on pre-trained linguistic patterns rather than entirely novel formulations (Ji et al ., 2023).”
Liu, T., Song, X., & Zhu, Q. (2026) · Humanities and Social Sciences Communications · doi:10.1057/s41599-025-06291-9 · Cited by 2
“…AI-generated content may contain inaccuracies — such as hallucinations (i.e., con fi dent but false outputs) or fabricated facts — a more reliable approach is to re fi ne and adapt existing, veri fi ed information rather than introducing unveri fi ed new claims (Ji et al., 2023; Liang, et al., 2024).”
Özergin, B. (2026) · Springer eBooks · doi:10.1007/978-3-658-46709-8_52-1
Towards Practical Ethics for AI
Dyoub, A., Lisi, F. A., Costantini, S., Letteri, I., Gasperis, G. D., & Gallese, C. (2026) · Intelligenza Artificiale · doi:10.1177/17248035251392741 · Cited by 3
“However, while LLMs offer powerful natural language processing, they inherently risk generating hallucinated or contextually incorrect information (Ji et al. (2023)).”
Revisiting data analysis with Pre-trained foundation models
Liang, C., Yang, D., Liang, Z., Liang, Z., Zhang, T., Xiao, B., Yang, Y., Wang, W., & Wang, H. (2026) · The VLDB Journal · doi:10.1007/s00778-025-00953-5 · Cited by 5
AI‐Assisted Bioelectronics for Personalized Health Management
Xiong, H., Dai, C., Chen, X., Kong, J., Wei, D., Fang, X., & Weng, W. (2026) · Advanced Electronic Materials · doi:10.1002/aelm.202500744 · Cited by 1
How Accurate Are the Responses of 4 AI Chatbots to Orthognathic Surgery Questions?
Hatipoğlu, Ş., Özkan, E. Ç., Taşova, F. A. K., & Zincir, Ö. Ö. (2026) · Journal of Craniofacial Surgery · doi:10.1097/scs.0000000000012367 · Cited by 1
Sreekanth, D., Gopi, S., & Dehbozorgi, N. (2025) · IEEE Frontiers in Education Conference (FIE) · doi:10.1109/fie63693.2025.11328160 · Cited by 4
“…to assess response quality—Exact Match for answer accuracy, Faithfulness and Hallucination for content reliability, Explanation Similarity, BERTScore, BLEU, and ROUGE for explanation quality, Context Recall for retrieval relevance, and Uncertainty for evaluating model reliability [48], [49].”
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.




