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 3,301–3,400 of 4,844
Editorial: Ethical considerations of large language models: challenges and best practices
Velmovitsky, P. E., Arbuckle, L., & Papadopoulou, P. (2026) · Frontiers in Digital Health · doi:10.3389/fdgth.2026.1807664
Jang, P.-S., & Lee, J.-Y. (2026) · The Journal of Korean Institute of Information Technology · doi:10.14801/jkiit.2026.24.2.169
RoboCritics: Enabling Reliable End-to-End LLM Robot Programming through Expert-Informed Critics
Kim, C. Y., White, N. T., He, E. C. L., Sala, F., & Mutlu, B. (2026) · ACM/IEEE International Conference on Human-Robot Interaction (HRI) · doi:10.1145/3757279.3785550 · Cited by 6
Toward Generalizable, General-Purpose, and Multimodal Knowledge Editing for Foundation Models
Wang, H., Liu, T., Ma, F., & Gao, J. (2025) · IEEE ... International Conference on Data Mining workshops · doi:10.1109/icdmw69685.2025.00304
“Without this ca-2477 pability, LLMs risk propagating outdated or even harmful information, especially in high-stakes domains [11, 13, 4].”
Infusing Structural Knowledge for Language Generation with Relational Graph Diffusion
Wang, S. H., Zhang, W., Zhu, J., Wang, Z., & Xia, Y. (2025) · IEEE ... International Conference on Data Mining workshops · doi:10.1109/icdmw69685.2025.00211
“However, most of them fuse sequences and KGs in a shallow manner [4] or indirectly exchange information between the two modalities [20], [21], leading to inaccurate answers or poor generalization ability [22], [23].”
E³-Guarded Generation: Provably Mitigating Hallucinations in Large Language Models
Wang, G., & Nixon, M. S. (2026) · 2012 International Conference on Computing, Networking and Communications (ICNC) · doi:10.1109/icnc68183.2026.11416906
“Despite their remarkable fluency and breadth, these systems exhibit a critical weakness that undermines their reliability: the tendency to generate hallucinations —confident assertions that are factually incorrect, internally inconsistent, or unsupported by available evidence [4].”
Automatic Department Performance Assessment via Strategic Prompt Engineering
Chen, L., Xu, S., & Qin, Y. (2026) · International Conference on Intelligent Control and Information Processing · doi:10.1109/icicip67436.2026.11417533
“[17] In serious scenarios such as performance evaluation, the model may misinterpret the meaning of text or generate non-existent ”arguments”.”
Community Member and Scientific Queries for Climate Resilience: An Assessment of the LLM Landscape
Nankabirwa, R., & Janeja, V. P. (2025) · IEEE ... International Conference on Data Mining workshops · doi:10.1109/icdmw69685.2025.00071
Corporate greenwashing accusations are selective signals of climate credibility gapsPreprint
Hu, E., Andreeva, G., Cojoianu, T., & Wang, S. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6205698
Abedi, M., Sanusi, I. T., & Tukiainen, M. (2026) · AI Brain and Child · doi:10.1007/s44436-026-00029-3
Liang, Q. Y., Zeng, Y., & Tang, M. (2026) · Scientific Reports · doi:10.1038/s41598-026-43529-1 · Cited by 5
A White-Box Benchmark for Intrinsic Hallucination Detection in Turkish LLMs
Bayramoglu, B., Ficici, M., Bildirici, F., Kızıltepe, R. S., & Karakuş, M. (2026) · International Informatics and Software Engineering Conference (IISEC) · doi:10.1109/iisec69317.2026.11418463 · Cited by 1
“This phenomenon, referred to as hallucination , remains a major barrier to the deployment of LLMs across diverse applications [2].”
SuperSAGA: A Supervisor-Subordinate Agentic workflow for the Generation of Assertions
Paul, S., Banerjee, A., Ghosh, S., Surendran, S., & Gajavelly, R. K. (2026) · Asia and South Pacific Design Automation Conference (ASP-DAC) · doi:10.1109/asp-dac66049.2026.11420235 · Cited by 2
“Nevertheless, a significant challenge in using LLMs for SVA generation is their tendency to produce plausible but inaccurate outputs [7] when handling sophisticated hardware protocols [8].”
SuSEF: Summarized Section Embedding with Fusion for Scientific Paper Recommendation
Westmoreland, J., & Liang, H. (2025) · International Conference on Robotics, Automation and Artificial Intelligence (RAAI) · doi:10.1109/raai67517.2025.11423046
“This has been shown to reduce hallucination and increase accuracy of LLM responses [10].”
Informed Skepticism: Navigating The Code Hallucination Dilemma
Faker, O., & Toker, Ş. (2026) · International Informatics and Software Engineering Conference (IISEC) · doi:10.1109/iisec69317.2026.11418436
“This problem happens because of the models’ statistical nature and training on bad data, or sometimes because they misunderstand a prompt [7][8].”
Key-augmented neural triggers for knowledge sharing
Wolf, A., Palma, M. E., Rani, P., & Gall, H. C. (2026) · Journal of Systems and Software · doi:10.1016/j.jss.2026.112850 · Cited by 1
Proposta per un inconscio collettivo artificiale
Gramantieri, R. (2026) · STUDI JUNGHIANI · doi:10.3280/jun62-2025oa20952
Artificial intelligence technologies in newsrooms: investigating Nepali journalists’ experiences
Pandey, L., & Ghimire, S. N. (2026) · SN Social Sciences · doi:10.1007/s43545-026-01398-5
“These concerns largely entailed unacknowledged AI use, inadequate human oversight, credibility, and accountability issues that could emanate from AI-induced hallucinations and intellectual laziness (Ji et al. 2023; Opdahl et al. 2023; van Dalen 2026).”
Semantic Web and Software Agents – A Forgotten Wave of Artificial Intelligence?
Pitkäranta, T., & Eero Hyvönen, A. (2026) · Lecture notes in computer science · doi:10.1007/978-981-95-5009-8_10 · Cited by 4
“One major concern is the potential for ”hallucinations,” where the LLM generates plausible-sounding but factually incorrect or nonsensical information (Ji et al., 2023).”
LLMS, Manufacturing Knowledge Graphs & GraphRAG enabling Intuitive Analytics
Senington, R., Mittermeier, L., & Ng, A. H. C. (2026) · IOP Conference Series Materials Science and Engineering · doi:10.1088/1757-899x/1342/1/012057
“Issues with LLMs are, however, also widely reported, where they fail to give correct results, such as is seen when they hallucinate [10, 11] and more specifically seen in a detailed analysis of their coding capabilities [12].”
Calado, S. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-4521-5.ch004 · Cited by 1
Chang, Y., Gao, T., & Li, Z. (2026) · Knowledge-Based Systems · doi:10.1016/j.knosys.2026.115765
Qi, Y., & Murat, A. (2026) · Journal of King Saud University - Computer and Information Sciences · doi:10.1007/s44443-026-00638-1 · Cited by 1
“However, probabilistic generation remains prone to hallucinations (Ji et al. 2023), which results in the production of content that is not grounded in the source text.”
A Geometric Taxonomy of Hallucinations in LLMsPreprint
Marin, J. (2026) · Qeios · doi:10.32388/qyjyru
Taranowski, A. O., & Samoylov, V. B. (2026) · Èlektronnoe modelirovanie · doi:10.15407/emodel.48.01.051
A comprehensive survey on table question answering: Datasets, methods and future directions
Xu, W., Liu, Y., Lu, L., Li, H., & Sun, G. (2026) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2026.114459 · Cited by 2
Rahman, M., & Chowdhury, M. (2026) · International Conference on Electrical, Computer & Telecommunication Engineering (ICECTE) · doi:10.1109/icecte69292.2026.11429437
“The work in [31] provides an overview of hallucinations in NLG by differentiating between intrinsic and extrinsic hallucinations.”
Anjum, U. B., Hirasawa, K., & Fatima, U. (2026) · AI and Ethics · doi:10.1007/s43681-026-01075-6 · Cited by 1
Axetorn, J., Edholm, F., Dobslaw, F., & Gren, L. (2026) · Requirements Engineering · doi:10.1007/s00766-026-00457-w · Cited by 1
“LLMs occasionally generate unfaithful or incorrect content, a phenomenon often called hallucination [27].”
Integrating Large Language Model Applications in Graduate Applied Behavior Analysis Education
Wilson, J. B., & Karsina, A. (2026) · Behavior Analysis in Practice · doi:10.1007/s40617-026-01159-z
“This phenomenon, known as an AI hallucination, occurs when the system generates information that appears credible but is entirely false or unsupported by existing sources, sometimes even fabricating journal titles, authors, or data that do not exist (Ji et al., 2023).”
Harnessing ChatGPT’s potential: A comprehensive guide for professors
DeRoy, T., Edmonds, M. A., & Messina, M. (2026) · Journal of Accounting Education · doi:10.1016/j.jaccedu.2026.101007 · Cited by 1
A bird-inspired artificial intelligence framework for advanced large text summarization
Huang, B., & Bari, A. (2026) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2026.1703769
Bauerdick, J., Streibel, L., Milde, M., et al. (2026) · Zeitschrift für wirtschaftlichen Fabrikbetrieb · doi:10.1515/zwf-2026-1010
A knowledge prompt augmented lightweight multimodal language assistant for biomedicine
Liu, L., Su, X., Zhou, X., & Gao, G. (2026) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2026.114516
From AI Stack to Cognitive Stack: A Semantic–Spatial–Relational Architecture for Trustworthy AI
Johnson, L., Inamdar, A. H., & Yadecha, B. L. (2026) · Trends in Computer Science and Information Technology · doi:10.17352/tcsit.000105 · Cited by 2
“Large language models generate responses through probabilistic token prediction rather than grounded reasoning, which can produce con fi dent, yet inaccurate outputs commonly described as hallucinations [4,5].”
Chen, Z., Du, M., Qian, C., & Yu, H. (2026) · Journal of Hazardous Materials · doi:10.1016/j.jhazmat.2026.141820 · Cited by 3
Generative AI–Driven Rumor Propagation Chains and the Dilemma of Legal Governance
Wu, W. (2026) · Journal of Education Humanities and Social Sciences · doi:10.54097/k7ngj392
Zhu, Y., Zhao, Z., Zhang, S., Sun, J., & Yu, K. (2025) · Globecom · doi:10.1109/globecom59602.2025.11431772 · Cited by 1
“Critically, prompt-only approaches are prone to the well-known hallucination problem [4] of large language models, which may result in unsafe or unrealistic command generation during action planning.”
HPC-MentorFlow: Strengthening Institutional HPC Competencies through an LLM-Enhanced Tutoring System
El-Amrany, S., Jarerattanachat, V., Brust, M. R., Pecero, J. E., Piyatumrong, A., & Bouvry, P. (2026) · International Conference on Knowledge and Smart Technology · doi:10.1109/kst67832.2026.11431975
SimEA: an entity alignment model based on multi-aspect similarity features
Zhao, W., Chen, Q., & You, J. (2026) · PeerJ Computer Science · doi:10.7717/peerj-cs.3729
Nurhidayati, H., Nurchayati, Z., & Sugiarto (2026) · Research Square · doi:10.21203/rs.3.rs-8994226/v1
Thankachan, B., Akkil, D., Rahman, S., Jokinen, K., & Turunen, M. (2026) · Multimodal Technologies and Interaction · doi:10.3390/mti10030031 · Cited by 1
Utsunomiya, T. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6316958
Li, X., Wang, S., Zhao, T., Shang, Z., & Yu, Z. (2026) · Information Processing & Management · doi:10.2139/ssrn.6447289
Leveraging Large Language Models in Engineering Design and Product Development: A Snapshot
Altun, O., Ott, M., Meihöfener, N., Budde, F., & Mozgova, I. (2026) · Procedia Computer Science · doi:10.1016/j.procs.2026.02.040 · Cited by 2
BioGraphletQA: Knowledge-Anchored Generation of Complex QA Datasets
Jonker, R. A. A., Martins, B. M. R. D. A., & Matos, S. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-21321-1_62 · Cited by 1
“However, despite their success, LLMs struggle with factual consistency, often generating hallucinated or inaccurate responses [12,11].”
Banwait, H. S., & Prabha, C. (2026) · International Conference on Image Processing and Capsule Networks (ICIPCN) · doi:10.1109/icipcn67432.2026.11438782
“Sometimes LLMs can provide us with fluent and grammatically correct results but results can be contextually wrong [7].”
Speculative Retrieval-Augmented Generation for Faithful Question Answering in Aircraft Maintenance
Signé, Q. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-21324-2_50
CancerRAGent: Evidence-Linked and Safety-Guided Oncology Question Answering
Vo, T., Trieu, A., Tran, V., Matsumoto, Y., & Nguyen, L.-M. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-21321-1_20
Building Trust in AI: A Qualitative Study of Human-AI Interaction
Leutheuser, V., Schäfer, F., Müller, J. M., & Voigt, K.-I. (2026) · Procedia Computer Science · doi:10.1016/j.procs.2026.02.048
A Call for Action: The Eloquence of GenAI Numbs Students’ Critical Thinking
Jafari, E., & Vassileva, J. (2026) · Lecture notes on data engineering and communications technologies · doi:10.1007/978-981-95-4423-3_43 · Cited by 2
AI vs. Human Text Detection: A High-Accuracy Ensemble Approach Using Machine Learning
Kökver, Y. (2026) · Bitlis Eren Üniversitesi Fen Bilimleri Dergisi · doi:10.17798/bitlisfen.1796956
Kim, S., Choi, H., & Huang, J. (2026) · Applied Sciences · doi:10.3390/app16073123
Luo, M. M., Zhang, P. M., Xu, W., Su, H., Zhou, X., Yan, H., & Cummings, J. J. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6464746
Sim, J.-A., Horan, M. R., Huang, X., Kim, M., Srivastava, D. K., Ness, K. K., Hudson, M. M., Baker, J. N., & Huang, I. (2026) · Communications Medicine · doi:10.1038/s43856-026-01499-5
Akkaya, A. (2026) · Akıllı Ulaşım Sistemleri ve Uygulamaları Dergisi · doi:10.51513/jitsa.1848120 · Cited by 1
Tuli, D., Yadav, H. C., & Rawat, A. (2025) · International Conference on Artificial Intelligence and Speech Technology (AIST) · doi:10.1109/aist68591.2025.11441547
“A. Large and Mid Language Models Large Language Models, such as GPT-3 [1], BLOOM [3], and Pythia [5], have demonstrated incredible natural language comprehension and generation abilities for a myriad of tasks.”
Hallucination-aware Retrieval Augmented Generation with Evidence-based Semantic Clustering
Jin, C., Lu, Y., Nie, J., & Yu, S. (2026) · International Conference on Artificial Intelligence and Social Network Systems (AISNS) · doi:10.1109/aisns67921.2026.11440643
“However, despite their impressive performance, a critical limitation persists: LLMs often generate hallucinations [3].”
Cheng, W., Zhu, H., & Lian, Y. (2026) · Value in Health · doi:10.1016/j.jval.2025.12.019
RetractedKeenan, M., Karachiwalla, N., Koo, J., Mwangi, C. W., Breisinger, C., & Kim, M. (2026) · PLoS ONE · doi:10.1371/journal.pone.0342852 · Cited by 2
“However, even with recent improvements, trust in AI remains a concern, with only 39% of people globally trusting AI-generated information since AI can make errors, be biased [7], spread misinformation [8], or make up information [9–11].”
Turbocharging Brazilian Mergers and Acquisitions: Questions & Answers Evaluation
Rubin, F. S., Moura, P. N. D. S., & Alvim, A. C. F. (2026) · Journal of the Brazilian Computer Society · doi:10.5753/jbcs.2026.5703
Bao, M., Cao, C., & Zhang, Y. (2025) · International Conference on Artificial Intelligence and Teacher Education (ICAITE) · doi:10.1109/icaite68636.2025.11442463
“Implementation Previous explorations have highlighted the risks associated with intelligent tools, such as hallucination and algorithmic bias during use [37, 38].”
Investigating the Impact of API Design Techniques for Tools in Agentic Applications
Bernardino, K. D., Queiroz, P. G. G., Guerra, E., Gonçalves, E., & Praxedes, L. B. F. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-39143-8_21
Peng, Y. L., Lee, W., & Juan, C.-Y. (2026) · Journal of Data and Information Quality · doi:10.1145/3802916 · Cited by 1
Enhanced multi-tuple extraction for materials: integrating pointer networks and augmented attention
Hei, M., Zhang, Z., Liu, Q., Pan, Y., Zhao, X., Peng, Y., Ye, Y., Zhang, X., & Bai, S. (2026) · Journal of Materials Informatics · doi:10.20517/jmi.2025.75 · Cited by 1
Diagnosing LLM Hallucinations in Process Mining Tasks: a Taxonomy and a Benchmark
Berti, A., & Kourani, H. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-17618-9_41
Improving AI Accuracy: Identifying and Fixing Hallucinations in Large Language Models
Dadoriya, N., & Chandra, S. K. (2025) · International Conference on Data Science and Intelligent Network Computing (ICDSINC) · doi:10.1109/icdsinc66221.2025.11448207
“Some algorithms used by AI models, such as statistical sequence pre-diction algorithms, do not process text with mean-ing, so these models generate ambiguous content that sometimes provides misinformation to the user [8].”
Güvenç, A. (2026) · Journal of Global Tourism and Technology Research · doi:10.54493/jgttr.1859320
Democratizing AI‐Enhanced Experimental Optimization: An Accessible Workflow for Chemists
Dahlen, N. N., & Forbes, T. Z. (2026) · Chemistry - Methods · doi:10.1002/cmtd.202600004
Wang, Y., & Fussell, S. R. (2026) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3799448 · Cited by 3
Trust-Aware Routing for Distributed Generative AI Inference at the Edge
Nguyen, C. L. T., & Elmroth, E. (2026) · International Conference on Distributed Computing in Smart Systems and the Internet of Things (DCOSS-IoT) · doi:10.1109/dcoss-iot69657.2026.00036
“Nevertheless, a substantial performance gap remains: reducing model scale often degrades complex reasoning ability, factual accuracy [16], and robustness against hallucinations [17], [18].”
Jayakody, A., Abeyrathne, O., Perera, T., et al. (2026) · International Conference on Advanced Research in Computing (ICARC) · doi:10.1109/icarc68737.2026.11453838
Hirumi, A. “. C., Giacumo, L. A., Kurzweil, D., Velasco, E. D. L. M., & Moon, H. (2026) · TechTrends · doi:10.1007/s11528-026-01177-2
“Regardless of the platform, transparency, reproducibility, and human verification remain essential, as AI outputs are prone to omissions, hallucinations, and inconsistent retrieval (Achiam et al., 2023; Bender et al., 2021; Ji et al., 2023).”
A Framework for Identifying Problem-Solution in Research Articles: Focusing on Social Issue
Kim, M.-S., Shin, H., Chun, H.-W., & Lee, J.-M. (2026) · IEEE Access · doi:10.1109/access.2026.3678862
“Despite this flexibility, LLM-based approaches often face challenges in domain-specific scientific applications, including limited reproducibility, difficulty in fine-tuning on custom-annotated datasets, and concerns about hallucination and explainability [6], [7].”
Digital Representation of NDE Systems: Data Networking and Information Modeling
Panchal, D., Leinenbach, F., Ardic, C. E., et al. (2026) · Applied Sciences · doi:10.3390/app16073447
“As a result, the domain-adapted LLM provides reduced hallucination tendencies [41] and delivers more precise and context-aware outputs.”
Russ, P. G., Bedenbender, S., Einloft, J., Meyer, H. L., Wenzel, L. T., Ganser, A., Hirsch, M. C., & Grgic, I. (2026) · Scientific Reports · doi:10.1038/s41598-026-46846-7 · Cited by 3
“At present, these systems remain susceptible to so-called hallucinations, defined as outputs that are factually incorrect, fabricated, or misleading yet presented in a coherent and persuasive manner 34,35 .”
Lee, W.-Y., Kim, J.-H., Leem, J., Lee, B.-W., Lee, S., & Kim, Y. W. (2026) · Applied Sciences · doi:10.3390/app16073377
Reversing the Death of Hypertext: Mechanistic Interpretability for LLM Navigation
Zheng, W., He, Y., & Li, J. (2026) · ACM SIGWEB Newsletter · doi:10.1145/3801080.3801084
AI-Driven Backlog Refinement for Enhancing User Story Quality and Sizing in Agile Projects
Rahmahsari, S., & Sekti, B. A. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-6851-1.ch008
EndoBot: An LLM Workflow-Driven Agent for Intelligent Gastrointestinal Endoscopy ConsultationPreprint
Zhou, S., Feng, B. C., Zhang, S., Zhang, W., Cui, Q., Qu, Y., Gao, J., Wang, H., Zhou, J., & Jiang, X. (2026) · Research Square · doi:10.21203/rs.3.rs-8510714/v1
PI-EnLLM: Personalized Interactive Healthcare Assistance via Ensembling Large Language Models
Dwibedi, A. K., Bhattacharjee, S. D., Chang, Y.-P., & Yuan, J. (2026) · Lecture notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering · doi:10.1007/978-3-032-16099-7_20
Kim, J., Kim, J. H., & Choi, A. (2026) · Computational and Structural Biotechnology Journal · doi:10.34133/csbj.0064
“The phenomenon of “hallucination”—generating factually incorrect information—is a critical risk in a clinical context [ 11 ].”
Amugongo, L. M., Schaller, L., Wendt, H., et al. (2026) · doi:10.2196/preprints.95417
Ganczarek, J., Stachów, W., & Leszczyński, M. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6512536
TVM-SimpleQA: A Factuality Evaluation Dataset for Traditional Vietnamese Medicine
Nguyen, G., Nguyen, T., Nguyen, L., & Đinh, Đ. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-21631-1_36
Agentic AI Meets RAG: A Framework for Domain-Specific Knowledge Evaluation
Vijay Kumar, K., M, A. A., Uma, K. V., & S, V. V. (2025) · IEEE International Conference on High Performance Computing, Data and Analytics Workshop (HiPCW) · doi:10.1109/hipcw66559.2025.00044
“Uniqueness = 1 − # semantically duplicate questions # total generated questions (5) 5) Factual Grounding: Grounding evaluates whether the system’s responses are factually supported by retrieved evidence, thereby minimizing hallucinations [23].”
Table-Aware Row-Level RAG for Classical Chinese Understanding
Liu, Z., & Leong, W. Y. (2026) · Computers · doi:10.3390/computers15040221
Reliable Batch-Sequential Model Editing via Enhanced Editing Anchor Compression
Xu, H., Peng, Z., Wang, X., Guo, H., & Ling, Z. (2026) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3803803
Academic Query Orchestration: An AI Agent Framework Utilizing RAG and Text-to-SQL
Vu, D.-H., Minh, D. V., & The, V. T. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-21625-0_6
Ha, O.-S. (2026) · Journal of Digital Contents Society · doi:10.9728/dcs.2026.27.3.821
LGA: lightweight design and privacy analysis of generative agents in social simulations
Guo, Y., Lu, X., Ma, W., Xiao, X., & Zhu, T. (2026) · International Journal of Information Security · doi:10.1007/s10207-026-01244-y
Zhang, Y., Shaik, F. A., Acharjee, S., Khalid, F., & Oussalah, M. (2026) · Reliability Engineering & System Safety · doi:10.1016/j.ress.2026.112674 · Cited by 18
“Furthermore, even after preference alignment, vision-language models may still produce unsupported causal reasoning when visual evidence is weak [87].”
Degachi, S., Niforatos, E., & Kortuem, G. (2026) · Behaviour and Information Technology · doi:10.1080/0144929x.2026.2638907 · Cited by 1
Huang, S. (2026) · ITM Web of Conferences · doi:10.1051/itmconf/20268403005 · Cited by 2
Explainable and accurate ENT diagnosis via a dual knowledge LLM
Gong, J., Yuan, H., & Li, F. (2026) · Health and Technology · doi:10.1007/s12553-026-01068-9
Agentic LLM-Driven Simulation Testing and Automated Analysis for Small Uncrewed Aerial SystemsPreprint
Garapati, J., Shah, A., Zhang, B., Vierhauser, M., & Agrawal, A. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6537456
Prakash, B. B., Nataraj, K. S., Shwetha, Thrupti, D. S., Kumar, J. A., & Tiwari, N. (2026) · IEEE International Conference on Interdisciplinary Approaches in Technology and Management for Social Innovation (IATMSI) · doi:10.1109/iatmsi68868.2026.11465869
“First, LLMs are known to hallucinate i.e., produce factually incorrect or misleading information with high confidence [11], [12].”
Detection of LLM Hallucinations Using Late Internal Representations
Hossan, S., & Deng, J. (2025) · International Conference on Machine Learning and Applications · doi:10.1109/icmla66185.2025.00214
“Ji et al. [10] and Rawte et al. [11] expanded this by categorizing hallucinations in LLM outputs — distinguishing types (e.g., intrinsic vs. extrinsic) and addressing evaluation gaps in benchmark datasets.”
A Critical Perspective on Adopting Generative AI in Marketing for Business Organizations
Zalan, T., & Toufaily, É. (2026) · Springer eBooks · doi:10.1007/978-3-031-75316-9_46-1
Erreurs d’orthographe chez les personnes dyslexiques : analyses et pistes de remédiation
Bodard, J. (2024) · theses.fr (ABES) · doi:10.70675/eb2c604czac70z4b15zb649zd1ceeca2b79c
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.




