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 3,701–3,800 of 4,842
He, S. T., Jiang, D., Jiang, R., Yin, M., Yu, S., Jiang, X., Zhou, L., Zhou, Z., Wei, H., & Xiao, J. (2026) · Brain and Spine · doi:10.1016/j.bas.2026.106109
Taysi, M. A., Yılmaz, R. E., Bozkurt, H., Özmen, A. İ., & İnce, G. (2026) · International Congress on Human-Computer Interaction, Optimization and Robotic Applications (ICHORA) · doi:10.1109/ichora69329.2026.11537178
“However, even when retrieval is relevant, a generator can still overstate what the evidence supports, blend weak signals into categorical claims, or propose actions whose rationale cannot be audited from cited evidence [3].”
When the Chain Breaks: Interactive Diagnosis of LLM Chain‐of‐Thought Reasoning Errors
Chen, S., Sritharan, N., Wen, X., Zhang, C., Wang, X., & Wang, Y. (2026) · Computer Graphics Forum · doi:10.1111/cgf.70439 · Cited by 2
Гамалій, К., Пересунько, Є., Борисенко, А., & Каук, В. (2026) · ГРААЛЬ НАУКИ · doi:10.36074/grail-of-science.15.05.2026.111
Ethical Implications of LLM Use in Sensitive Clinical Domains
Tyagi, V., & Sharma, S. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-2600-2173-6.ch006
АНАЛІТИЧНИЙ ЧАТ ДЛЯ АНАЛІЗУ ТАБЛИЧНИХ ДАНИХ НА ОСНОВІ LLM ТА MCP
Bilova, T., Землякова, О., & Панкратович, А. (2026) · ГРААЛЬ НАУКИ · doi:10.36074/grail-of-science.15.05.2026.118
Stability-Aware and Phase-Aware Identification of Hallucination-Associated Neurons in LLMs
Xu, Z., Yang, C., Zhao, L., & Gao, Y. (2026) · International Conference on Intelligent Computing and Signal Processing (ICSP) · doi:10.1109/icsp69961.2026.11540612
“Hallucination has been framed as a general failure mode of neural text generation rather than a pathology of any single model family [1].”
Effects of AI Hallucination Disclosure on User Trust: Moderating Role of Virtual Human Interfaces
Kang, J. (2026) · Journal of Digital Contents Society · doi:10.9728/dcs.2026.27.5.1191
Ringwald, C., Gandon, F. L., Zucker, C. F., Michel, F., & Akl, H. A. (2026) · Semantic Web · doi:10.1177/22104968261452987
“This is a good starting point because language model hallucinations generally affect these literal values (Ji et al., 2023).”
Cha, H., Shin, S. J., Sim, J.-A., Yu, J., Chang, H., & Kim, J. (2026) · BMC Medical Ethics · doi:10.1186/s12910-026-01513-4
A Comprehensive Study on Evaluation Methods of Large Language Models
Suri, H., Bhardwaj, M. K., Pandey, B., et al. (2026) · doi:10.1109/icisessc68634.2026.11542722
“Several studies have shown that semantic similarity metrics to have strong correlation with human evaluations than the BLEU-based ratings [1],[6].”
RuleDroid: LLM-Augmented Synthesis of Static Security Detection Rules for Android Apps
Xie, Z., Chen, M., Gao, Y., Yang, S., Diao, W., Liu, X., & Zhang, K. (2026) · IEEE Transactions on Software Engineering · doi:10.1109/tse.2026.3700217
“LLMs also suffer from the “hallucination” problem [12], wherein they may produce answers without any relevant information, relying solely on existing knowledge.”
Luo, Y. (2026) · Research Square · doi:10.21203/rs.3.rs-9663460/v1
Towards higher quality and fewer hallucinations: A multi-agent collaboration framework for LLMs
Shen, S., Sun, D., Qin, Z., Su, Y., Zhu, L., Lu, J., Huang, Z., & Wang, S. (2026) · Information Processing & Management · doi:10.1016/j.ipm.2026.104953 · Cited by 3
Retrieval-Augmented Generation: Future of the Large Language Models?
Sanjay, G. S., You, W., & Ryu, J. (2026) · Journal of Sciences and Innovation for Sustainable Peace · doi:10.23184/jsisp.vol.4.no.1.08
Short Paper: Models in the Dark – Rectification and Erasure Under GDPR in ML Supply Chains
Graßhoff, H., Hansen, M., Jensen, M., & Ramezanian, S. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-35899-8_4 · Cited by 1
Damastuti, F. A., Gumelar, A. B., & Firmansyah, K. (2026) · Engineering Technology & Applied Science Research · doi:10.48084/etasr.17900
“With the rise of Large Language Models (LLMs), question generation quality has improved substantially [4-6].”
ETFAgents: A Multi-agent System with a Single LoRA-fine-tuned Agent
Pan, Y., Liu, X., & Chu, C. (2026) · International Journal of Computational Intelligence Systems · doi:10.1007/s44196-026-01401-0
“However, recent studies show that multi-agent prompting can improve reasoning and factuality only as a soft constraint [31], and that LLMs remain prone to hallucinations that can propagate within agentic workflows [32, 33].”
R3-RAG: Reweight, rerank, and reflect for evidence-calibrated scientific multimodal reasoning
Yin, Z., Wang, T., Qian, Y., Wang, K., Hu, P., & (8200), L. Z. (2026) · Expert Systems with Applications · doi:10.1016/j.eswa.2026.133104 · Cited by 2
Xiao, H., Zhu, T., Wang, L., & Xu, W. (2026) · Information Processing & Management · doi:10.1016/j.ipm.2026.104949
Hallucinated dosing: An emerging medication safety risk in generative AI
Zavaleta‐Monestel, E., Arguedas‐Chacón, S., & Mora-Jiménez, J. (2026) · American Journal of Health-System Pharmacy · doi:10.1093/ajhp/zxag177
Entity Guided Natural Language Generation for Targeted Review Summarization
Prachi, P., Malavika, T., Mahendran, P., et al. (2026) · International Conference on Pervasive Computing and Social Networking (ICPCSN) · doi:10.1109/icpcsn68523.2026.11543966
Generative AI in Intelligent Tutoring Systems: Paradigm Shift and Hybrid Control
Rida, Z., Boukachour, H., Ennaji, M., & Machkour, M. (2026) · International Conference on Innovative Research in Applied Science, Engineering and Technology (IRASET) · doi:10.1109/iraset68627.2026.11538851
“The most critical challenge is the tendency of LLMs to “hallucinate,” i.e., to generate factually incorrect but gram-matically plausible information [6].”
Abdelali, E. A. E. M., Salma, E.-M., Sabri, M. A., & Yahyaouy, A. (2026) · International Conference on Innovative Research in Applied Science, Engineering and Technology (IRASET) · doi:10.1109/iraset68627.2026.11538759
“Training data is grounded in verified documents following RAG-inspired principles [24], reducing hallucination [23].”
InfluentialLiang, W., Yang, J., Li, S., Luo, P., Deng, H., & Li, T. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6902284
Intelligent Tutoring Systems from Symbolic AI to LLMs: A Critical Review
Salh, B., Bouskour, S., & Oughdir, L. (2026) · International Conference on Innovative Research in Applied Science, Engineering and Technology (IRASET) · doi:10.1109/iraset68627.2026.11538663
“[45] and pedagogical inconsistencies, multi-agent architectures are emerging with specialized agents collaborating on learner modeling, content generation, and evaluation [5],[46].”
RHLMA: A Role-Hierarchical LLM Multi-Agent Framework for Airborne Software Requirement Review
Sun, J., Yang, R., Gu, J., Chen, L., Zhao, Y., & Yang, X. (2026) · International Conference on Multi-scale Artificial Intelligence (MAI) · doi:10.1109/mai69289.2026.11543713
“Large language models often generate erroneous judgments with unjustifiably high certainty and no traceable evidence—hallucination or miscalibration [7]—which severely compromises auditability and certification compliance in safety-critical domains [8], rendering review results unreliable and…”
El Kayali, M. K., Neumann, E. M., Fahy, S., Bartek, B., Oehme, S., Getgood, A., Jung, T., & Milinkovic, D. D. (2026) · Journal of ISAKOS · doi:10.1016/j.jisako.2026.101160
Li, Y. Z., Wang, X., Liu, R., Wang, M., & Luo, X. (2026) · International Conference on Information Science, Electrical and Automation Engineering (ISEAE) · doi:10.1109/iseae69422.2026.11544327
“Although LLMs excel at generating fluent dialogue [1], their reliance on a limited internal knowledge base makes them prone to generating outputs that contradict conversational history or external facts [2, 3].”
CODE ACROSTIC: Robust Watermarking for Code Generation
Li, L., Xin, S., Cao, Y., & Cao, X. (2026) · Lecture notes in computer science · doi:10.1007/978-981-95-7251-9_35
Mapping AI feedback under emotional prompts: a sentiment-topic-network approach
xiangming, L., & Wei, W. (2026) · International Journal of Information and Learning Technology · doi:10.1108/ijilt-04-2025-0121 · Cited by 2
Lin, B., Mao, S., Wang, P., Zhang, L., & Haosu, Z. (2026) · International Conference on Information Science, Electrical and Automation Engineering (ISEAE) · doi:10.1109/iseae69422.2026.11544810
“However, the stochastic nature of LLMs leads to hallucinations, where the model may generate tactically plausible but legally prohibited actions, severely violating ROE constraints [9].”
Li, Y., Fang, H., Ganz, K. J., Moskal, L. M., & Guo, Q. (2026) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2026.3701907 · Cited by 1
“Transformers have demonstrated remarkable performance in modeling sequential data [46], [47].”
Indexação Semântica Modular para Recuperação Colaborativa de Conhecimento em Sistemas RAG
Freire, E. C., de Azevedo, K. S., Silva, S. N., & Fernandes, M. A. C. (2026) · Simpósio Brasileiro de Sistemas Colaborativos (SBSC) · doi:10.5753/sbsc.2026.20073
Li, Y., Zhang, T., & Zhuang, W. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6907642
VERITAS: A Multi-Agent Verification-Based Framework for Reliable Intelligent Tutoring Systems
Victorio, L. S., de Araujo, A. V., Borges, V. A., Turine, M. A. S., & Nogueira, B. M. (2026) · ACM Symposium on Applied Computing · doi:10.1145/3748522.3779997
Dual-Driven AI for Eddy Current Array-Based Surface Defect Tomography
Chen, R., Gao, B., Chang, C., et al. (2025) · IEEE Far East NDT New Technology & Application Forum (FENDT) · doi:10.1109/fendt66689.2025.11547370
“Artificial intelligence (AI) is now pervasive, with applications spanning autonomous driving [1], natural language generation [2], and nondestructive testing (NDT) [3].”
A Theory Repair Based Traffic Regulations Generalisation for Autonomous Vehicles
Lu, Y., Li, X., Yu, Z., Lin, Y., Schafer, B., Bundy, A., Ireland, A., & Lin, Z. (2026) · Communications in computer and information science · doi:10.1007/978-981-95-8262-4_18
Safety Assurance Through Red-Teaming, MLOps, and Continuous Monitoring for Medical LLMs
Pandey, A. K., Sharma, S., Kumar, R., & Kumar, G. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-7862-6.ch014
Wang, S., & Yin, Z. (2026) · Expert Systems with Applications · doi:10.1016/j.eswa.2026.133113 · Cited by 1
Ensuring Safe and Ethical Use of LLMs in Healthcare
Kamaladevi, M., & S., R. M. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-7862-6.ch010
Valero‐Chillerón, M. J., Llagostera‐Reverter, I., Luna-Aleixós, D., & González‐Chordá, V. M. (2026) · Enfermería Clínica (English Edition) · doi:10.1016/j.enfcle.2026.502553
Toward Generative AI Applications in Disaster Management: A Critical Review and Conceptual Framework
Turan, M. (2026) · Sustainable development · doi:10.5772/intechopen.1014952
Evaluating teacher-supervised small medical language models for institution-constrained clinical QA
Du, G., Dimitriadis, D., & Li, X. (2026) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2026.105095
End-to-End Chatbot Evaluation with Adaptive Reasoning and Uncertainty Filtering
Dang, N., Le, T., & Nguyen, H. T. (2026) · Communications in computer and information science · doi:10.1007/978-981-92-0068-9_9
Phan, K., Le, X.-B., & Quan, T. (2026) · Lecture notes in computer science · doi:10.1007/978-981-92-0071-9_16 · Cited by 1
Kucharski, M. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-27927-9_23
Neris, L. M. (2026) · RCMOS - Revista Científica Multidisciplinar O Saber · doi:10.51473/rcmos.v1i1.2026.2477
Design and Implementation of a Generative Pre-Trained Transformer for Universities of Bangladesh
Ohi, M., Sajid, A., Akibuzzaman, P., Shahriar, S. M. N., Zaman, M., Islam, H. M. A., & Maharaz, M. M. (2026) · IEEE International Conference on Quantum Photonics, Artificial Intelligence & Networking (QPAIN) · doi:10.1109/qpain69676.2026.11545594
“[21], [22] LLMs with safety filters Risks exposing sensitive data.”
From Meetings to Product Vision: An Agentic Approach with MARARE
Sami, M. A., Evangelista, G., Kemell, K., et al. (2026) · IEEE Software · doi:10.1109/ms.2026.3702383
Lai, T., Liu, F., Li, G., & Hua, L. (2026) · Viruses · doi:10.3390/v18060656
Alsayyed, R., Khelfa, S., & Hawash, A. (2026) · ASU International Conference in Emerging Technologies for Sustainability and Intelligent Systems (ICETSIS) · doi:10.1109/icetsis68266.2026.11549183
“This implies that there is no guarantee that the model is grounded in verifiable information, and errors may still occur [5].”
AI-Driven Software Pricing: An Integrated Approach with Prompt Engineering for Market Analysis
Muniz, G. F., Tonera, J. D. C., Noll, R. P., & de Islabão, G. I. (2026) · Journal of the Brazilian Computer Society · doi:10.5753/jbcs.2026.6572
“However, limitations regarding hallucinations and reasoning can lead to inconsistent or incorrect responses [Ji et al ., 2023].”
Rathnayake, D. S., & Slotkienė, A. (2026) · New Trends in Computer Sciences · doi:10.3846/ntcs.2026.27121
Implicit Affective Steering: A Design Probe for Generative Curatorial Storytelling
Ma, M. C., Shi, L., Li, H., Cheng, J., & Kirk, D. S. (2026) · Designing Interactive Systems Conference · doi:10.1145/3800645.3812861
32715 Content analysis in quantitative communication research
Andreeva, I. N., Taggart, M., Lattimer, T. A., & Hahn, L. (2026) · De Gruyter eBooks · doi:10.1515/9783111087139-015 · Cited by 1
ADEPT: Interactive Visual Analytics for Audio Dataset Exploration and Preparation
Chen, C., Lin, T., & Kimball, J. (2026) · Designing Interactive Systems Conference · doi:10.1145/3800645.3812897
Tuxanbayev, A., Yrymbayeva, N., Suleymen, B., Zhanbyrbaevna, T. M., Rakhimgaliyeva, P., & Serikbayeva, N. B. (2026) · Cogent Education · doi:10.1080/2331186x.2026.2684322 · Cited by 3
Hopeful Failure: How Collaborative Design Fiction Reimagines AI
Basoah, J., Reinecke, K., Rosner, D. K., & Ogbonnaya-Ogburu, I. F. (2026) · Designing Interactive Systems Conference · doi:10.1145/3800645.3813092
“…framing, alignment extends beyond the accurate identification and execution of predefined user goals or values, and instead becomes a dynamic, context-dependent process that must be revisited as conditions change and systems exhibit known limitations such as misinformation and hallucinations [56].”
[R.H.J.M.K. and S.M.H. contributed equally to this work.]
Zöller, N., Berger, J., Lin, I., Fu, N., Komarneni, J., Barabucci, G., Laskowski, K., Shia, V., Harack, B., Chu, E., et al. (2025) · MPG.PuRe (Max Planck Society)
The Case for Repeatable, Open, and Expert-Grounded Hallucination Benchmarks in Large Language Models
Norman, J., Rivera, M. U., & Hughes, D. A. (2026) · International Conference on Information and Computer Technologies (ICICT) · doi:10.1145/3803291.3803328 · Cited by 1
“While there is a considerable body of work surrounding language model hallucination, the field has not settled on a single operational definition of what, specifically constitutes a hallucination [see e.g., 13, 17, for a survey of the field].”
TransFolk: Transformer-Based Generation of Folk Melodies
Martínez–Rodríguez, B. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-27827-2_38
Evaluating LLMs for Explaining Insecure Code and Identifying Vulnerabilities in Java and Python
Mahmud, A., Rawajfih, Y., & Wu, F. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-22211-4_12
Liu, S., & Yang, X. (2026) · Communications in computer and information science · doi:10.1007/978-981-95-9884-7_3
SoK: A taxonomy of LLM threats
Stylianou, I., Bountakas, P., Zarras, A., Farao, A., Bolgouras, V., & Xenakis, C. (2026) · Computer Science Review · doi:10.1016/j.cosrev.2026.101013 · Cited by 2
Mitigating Hallucinations in Vision-Language Models via Contextual Entropy Calibration
Yin, X., Peng, D., & Wei, W. (2026) · International Conference on Multimedia Retrieval · doi:10.1145/3805622.3810847 · Cited by 2
Arslan, O., & Giray, B. (2026) · Revista da Associação Médica Brasileira · doi:10.1590/1806-9282.20251769
“Large language models, including ChatGPT, have been shown to generate plausible but incorrect or misleading information, commonly referred to as hallucinations 18 .”
Ayalew, A. M., Hasan, M. R., Seppänen, T., & Oussalah, M. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-28819-6_38
On Possibilities and Challenges of GPT‐Assisted Learning Environments
Harnisch, M., & Bartel, T. (2026) · PAMM · doi:10.1002/pamm.70160
Martins, A., Lapão, L. V., Nunes, I. L., & Londral, A. R. M. (2026) · IFIP advances in information and communication technology · doi:10.1007/978-3-032-28939-1_11
Foundation Models: A Disruptive AI Paradigm
Schmid, T., Hildesheim, W., & Holoyad, T. (2026) · Springer eBooks · doi:10.1007/978-3-032-00533-5_4
Risk-Aware Medical Question Answering as a Reliability-Critical Decision Process
Luo, Y., & Li, Y. (2026) · International Conference on Multimedia Retrieval · doi:10.1145/3805622.3810784
Modelfile Engineering for Offline B2 English Tutoring
Techadisai, K., & Sirinaovakul, B. (2026) · doi:10.1145/3816713.3819514
“Ji et al. [11] surveyed hallucination reduction techniques in LLMs, highlighting system prompt design as critical for behavioral control.”
Liang, Z., Luo, B., Zhang, X., Zou, Y., & Ke, N. (2026) · International Seminar on Artificial Intelligence, Networking and Information Technology (AINIT) · doi:10.1109/ainit70033.2026.11557875
“LLMs tend to produce hallucinations, generating plausible yet factually incorrect content [3].”
ACTS A multi-tier benchmark evaluating LLM cipher identification under controlled blind conditions
Ibrahim, Y. H. Z., & Salama, M. K. (2026) · Expert Systems with Applications · doi:10.1016/j.eswa.2026.133329
“At 700 and 7,000 files, the same conclusion is supported by a deterministic heuristic proxy, whose stability indicates that the apparent reasoning trace is, in this task, confabulation [29, 30, 45]—sound-sounding analysis that is causally disconnected from the decision.”
Ethical issues of large language models: a multi-level thematic synthesis of the academic literature
Pervez, H., Minkkinen, M., Jylhä, H., & Mäntymäki, M. (2026) · AI and Ethics · doi:10.1007/s43681-026-01173-5 · Cited by 5
Resource-efficient retrieval-augmented question answering for the Indian Lok Sabha dataset
Sivakumar, S., & Meenakshi, S. P. (2026) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2026.1798021
Who Guards the Agents? A Control Plane Architecture for Responsible Agentic AIPreprint
Colombo, M. C., Paris, D., Aspromonte, M., Ermellino, A., Mombelli, N., Portanti, M., Zoccoli, A., Donelli, N., Castelnovo, A., & Cosentini, A. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6955958
Attention-Based Retrieval Quality Evaluation with Comparative Query Rewriting for Corrective RAG
Yuan, M., Li, Q., Yang, X., & Wang, H. (2026) · International Seminar on Artificial Intelligence, Networking and Information Technology (AINIT) · doi:10.1109/ainit70033.2026.11557873
“Large language models have demonstrated strong performance across a broad range of language tasks, yet their knowledge is fixed at training time and degrades on queries concerning recent events or long-tail entities [1].”
Xu, Z., Wang, Z., Wang, H., Zhou, Y., & Sun, M. (2026) · International Seminar on Artificial Intelligence, Networking and Information Technology (AINIT) · doi:10.1109/ainit70033.2026.11558060
“Such hallucination problems reduce the trustworthiness of the system and limit its practical value in simulation-based training assistance [5].”
Implicit to Explicit: A Pipeline for Inter-Policy Conflict Detection in Data Ecosystems Governance
Hmimou, A., Jovanović, P., & Romero, O. (2026) · IEEE International Conference on Data Engineering Workshops (ICDEW) · doi:10.1109/icdew71238.2026.00025
“However, because LLMs are fundamentally probabilistic, they are prone to hallucinations and non-deterministic outputs, making them unsuitable as direct enforcers for rigorous data ecosystem governance [10].”
OrthoEdit: Principled and Stable Knowledge Editing via Orthogonal Subspace Projection
Qiao, S., Liu, X., Gupta, A., & Na, S. (2026) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl.a.700
Медианные значения метрик соответствия для классов SUPORT и CONTRADICT
Кузнецов, И. (2026) · Cifra LLC Journals · doi:10.60797/irj.2026.168.94.4
Кузнецов, И. (2026) · Международный научно-исследовательский журнал · doi:10.60797/irj.2026.168.94
Stenzl, A., Rogers, E., Ananiadou, S., Wang, Y., Armstrong, A. J., Sboner, A., Cacciamani, G. E., Schijvenaars, B. J. A., Ahmed, K. R., Thomsen, H., et al. (2026) · JAMIA Open · doi:10.1093/jamiaopen/ooag078 · Cited by 1
“30 Sycophancy in LLMs which represents alignment with user inputs (containing potential inaccuracies) over accuracy of outputs was not assessed in this study.”
Rita, G. A. F., Melo, L. G. M. D. C., & Moreira, M. L. D. L. (2026) · RECIMA21 - Revista Científica Multidisciplinar · doi:10.47820/recima21.v7i6.8322 · Cited by 2
LLM and Mobile App-based Intelligent Campus Navigation System
Kim, H.-S., Jang, M.-G., Hong, S.-Y., Han, S. W., Kim, H.-M., & Lee, D. (2026) · The Journal of Korean Institute of Information Technology · doi:10.14801/jkiit.2026.24.5.189
Teng, J., & Cho, S. (2026) · Interactive Learning Environments · doi:10.1080/10494820.2026.2676667
Основные этапы комплексной оценки соответствия научных утверждений информации из первоисточника
Кузнецов, И. (2026) · Cifra LLC Journals · doi:10.60797/irj.2026.168.94.3
Современные методы повышения качества цитатно-осведомленной суммаризации
Кузнецов, И. (2026) · Cifra LLC Journals · doi:10.60797/irj.2026.168.94.2
Основные этапы цитатно-осведомленной суммаризации
Кузнецов, И. (2026) · Cifra LLC Journals · doi:10.60797/irj.2026.168.94.1
Style hybride d’une chimère. Authentifier le style dans une production textuelle hybride.
Maurel, P. (2026) · AI-Linguistica Linguistic Studies on AI-Generated Texts and Discourses · doi:10.62408/ai-ling.v4i1.37
Developing a Proof-of-Concept Automarker for Handwritten Mathematics Assessments
Mahlaola, N. M., & Martin, M. Y. (2026) · International Conference on Information and Education Technology (ICIET) · doi:10.1109/iciet69664.2026.11561567
“The largest contributor was the LLM Assessment stage (34%), where errors were primarily driven by output formatting inconsistencies and hallucinations [25].”
Automated quality evaluation of personalized marketing prompts generated by large language models
Jamshidi, S., Aik, M., Smith, J., et al. (2026) · Strategic Business Research · doi:10.1016/j.sbr.2026.100203
Upskilling UX Designers for AI-Native Work: A Pedagogical Framework and Empirical Evaluation
Álvarez, I. (2026) · Annual Symposium on Human-Computer Interaction for Work (CHIWORK) · doi:10.1145/3808045.3808059
Nguyen, T., Dang, T., Nguyen, D., & Vo, C. (2026) · International Conference on Information and Education Technology (ICIET) · doi:10.1109/iciet69664.2026.11561634
“More discussion on hallucination in natural language generation has been given in [9].”
Bamil, V. (2026) · IGI Global eBooks · doi:10.4018/979-8-3373-1509-6.ch003
Prayanto, F., & Perwira, R. I. (2026) · Digital System and Computing · doi:10.58920/dsc0201633
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.




