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 3,801–3,900 of 4,844
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
Roh, S.-H., & Kimb, T.-S. (2026) · Expert Systems with Applications · doi:10.1016/j.eswa.2026.133328
SlideSAVR: Enabling Live Analysis during Data Presentations via Multimodal Sketching and Voice Input
Han, C., Tanjim, M. M., Guo, S., Dierk, C., Isaacs, K. E., & Hoffswell, J. (2026) · Computer Graphics Forum · doi:10.1111/cgf.70459 · Cited by 2
Alshehhi, A. M., & Zakaria, K. A. I. (2026) · International Conference on Integrated Intelligence and Cognitive Engineering (ICIICE) · doi:10.1109/iciice69672.2026.11565268
“Prior work often separates hallucinations into intrinsic hallucinations, which contradict the provided context, and extrinsic hallucinations, which introduce unsupported content beyond the source material [1], [2].”
Pareek, C. S. (2026) · International Conference on Integrated Intelligence and Cognitive Engineering (ICIICE) · doi:10.1109/iciice69672.2026.11565486
“These dangers include the intrusion of privacy of information, partisanship in processing claims, and the unwanted formation of misleading or false information [3].”
Aggarwal, A. (2026) · International Conference on Integrated Intelligence and Cognitive Engineering (ICIICE) · doi:10.1109/iciice69672.2026.11565374
“GenAI: requires scenario libraries, rubric-based scoring, stress testing, and policy verification [9], [10].”
InfluentialJinna, K. R., Ponnoju, S. C., Namperumal, G., Thilagaraj, S., Kondisetty, K., & Srinivasan, V. (2026) · International Conference Inventive Material Science Applications · doi:10.1109/icima68728.2026.11564099
“Random Generation (RAG) frameworks [3] that can access external knowledge repositories at inference time, augment neural retrievers with generative models instead of relying exclusively on parameters [4].”
Досанова, Б. К., Akhmetova, A., & Akhmetova, Z. (2026) · InterConf · doi:10.51582/interconf.19-20.05.2026.002
Navigating maritime emergencies with large models: A lifecycle-oriented review
Xu, J., Yang, Q., Yang, Z., Gao, Y., Yang, K., & Jiang, J. (2026) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2026.115411 · Cited by 2
Adaptive air traffic management via large language model augmented reinforcement learning
Chen, S., Tabrizian, A., Wei, P., Wang, Y., & Qu, X. (2026) · Transportation Research Part E Logistics and Transportation Review · doi:10.1016/j.tre.2026.105027
Zhang, X., & Lee, J. (2026) · International Journal of Human-Computer Interaction · doi:10.1080/10447318.2026.2690096
AlDahoul, N., Karim, H. B. A., & Tan, M. J. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6900538
Clause Encounters of the Third Kind: Can LLMs Replace Language Teachers?
Šekrst, K., & Kovačić, A. (2026) · Oxford University Press eBooks · doi:10.1093/9780198945215.003.0252
“18.”
Blanchard, S. J., Garvey, A. M., & O'Laughlin, L. (2026) · Current Opinion in Psychology · doi:10.1016/j.copsyc.2026.102427
Warren, G., Sun, J., Shklovski, I., & Augenstein, I. (2026) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3805689.3812358 · Cited by 4
Ou, M., Chen, L., & Zheng, H. (2026) · Journal of Broadcasting & Electronic Media · doi:10.1080/08838151.2026.2692544 · Cited by 1
CONAN: CONcept-Oriented ANomaly Detector for Interpretable Zero-Shot Detection
Lee, Y., Lee, E., Kwon, J., Lee, S., Kim, K., Yu, S., & Kim, Y. (2026) · Lecture notes in computer science · doi:10.1007/978-981-92-1947-6_44
Demir, N. Y. (2026) · Digital Scholarship in the Humanities · doi:10.1093/llc/fqag084 · Cited by 1
GeoMind: An Autonomous Multi-Agent Framework for Verified Geospatial AnalysisPreprint
Habib, J., Srabon, M. I., Nahar, M. L., Lutfy, M., Khatun, M. A., Shammo, M. B. A. Z., & Hasan, M. A. (2026) · Research Square · doi:10.21203/rs.3.rs-10034367/v1
Ethics, Privacy, and Governance in AI-Enabled Information Systems
Shivaraj, G. (2026) · IGI Global eBooks · doi:10.4018/979-8-2600-0111-0.ch009
A Framework for Automated Cyber Threat Intelligence Extraction Based on Knowledge Graph
Liu, X., Tian, Y., Tang, L., & Zeng, S. (2026) · Chinese Control and Decision Conference (CCDC) · doi:10.1109/ccdc69976.2026.11559654
“This issue is further exacerbated by a lack of domain-specific knowledge and can be exploited to launch data poisoning attacks that spread false threat intelligence [10].”
MisEdu-RAG: A Misconception-Aware Dual-Hypergraph RAG for Novice Math Teachers
Guo, Z., Xue, R., Lu, Y., & Lin, J. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-29755-6_10
Shrestha, R. (2026) · International Conference on Artificial Intelligence and Industrial Technology Applications (AIITA) · doi:10.1109/aiita69518.2026.11567271
“[4,5] To move beyond simple accuracy, we evaluated robustness across different manual types using the RAGS framework; these findings are summarized in Table 2 .”
Dewi, A. N., Sørensen, B., & Thellefsen, M. (2026) · Journal of Documentation · doi:10.1108/jd-04-2026-0185
Alnazawi, A., & Almarhabi, M. (2026) · Advances in Simulation · doi:10.1186/s41077-026-00447-6 · Cited by 3
Scalable Personalized Education Through Agent-orchestrated Retrieval-Augmented Learning
Dr. Annapoorani, S., Shobanadevi, A., & Babu, S. (2026) · International Conference on Computational Robotics, Testing and Engineering Evaluation (ICCRTEE) · doi:10.1109/iccrtee68719.2026.11566356
“While recent advances in large language models offer powerful capabilities for content generation and natural language interaction, their direct use in education raises concerns around hallucinations, curriculum misalignment, and the lack of structured pedagogical reasoning [3].”
Lin, B., Yi, W., Haosu, Z., Wang, P., & Mao, S. (2026) · International Conference on Sensor-Cloud and Edge Computing System (SCECS) · doi:10.1109/scecs69812.2026.11565827
“While LLMs offer significant cognitive reasoning potential for situation assessment, their deployment at the tactical edge is strictly constrained by hardware limits and the risk of safety-critical logic hallucinations [3].”
AgentCite: Task-Isolated Multi-Agent Collaboration for Cost-Effective Long-Context Citations
Cao, Z., Wang, H., Jin, L., Zhu, T., & Ren, L. (2026) · International Symposium on Autonomous Systems (ISAS) · doi:10.1109/icaisisas68969.2026.11567713
“This lack of transparency makes it challenging for users to verify the outputs, exacerbating concerns about the reliability and potential hallucinations of LLMs [6].”
LLM-Based Multimodal Entity Linking
Liu, Q., Qiu, B., Xia, X., Sun, H., Zhou, Z., Yi, E., Ren, T., & Li, X. (2026) · Chinese Control and Decision Conference (CCDC) · doi:10.1109/ccdc69976.2026.11560337
“However, generative methods may suffer from hallucination, producing entities that do not exist in the knowledge base [20].”
Selyshcheva, I. A. (2026) · CTE Workshop Proceedings · doi:10.55056/cte.1438
“Surveys of the phenomenon distinguish, at the most general level, between intrinsic hallucinations that contradict a provided source and extrinsic hallucinations that cannot be grounded in any source at all [ 28 ] .”
Two-stage distributionally robust decoding for reliability-oriented natural language generation
He, G., La, L., Gao, K., & Zheng, J. (2026) · Information Processing & Management · doi:10.1016/j.ipm.2026.105000
Lee, T., Lee, S., & Kim, S. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-29791-4_30
A Systematic Evaluation of Decision Mechanisms in Multi-agent LLMs for Misinformation Detection
Lee, Y., Huang, Y., & Chu, T.-H. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-30044-7_6
Kiang, Y. W., Li, Z., Lee, A. T. L., Tam, V. W. L., Lam, K. H., & Lee, V. C. S. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-29791-4_26
QQJ: Quantifying Qualitative Judgment for Scalable and Human-Aligned Evaluation of Generative AI
Veysi, M., Shamsinejadbabaki, P., Zare, M., & Sabouri, M. (2026) · International Biennial Conference on Advances in Artificial Intelligence and Data Science (IBAAIDS) · doi:10.1109/ibaaids66771.2026.11567442
Lee, S., Jeon, K., Lim, J. S., Park, Y., Kim, M. S., Jin, J. H., Han, C. H., Yoon, S. H., & You, S. C. (2026) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2026.105071 · Cited by 1
Wu, H., Cui, Q., Wang, J., Shi, L., & Li, W. (2026) · Expert Systems with Applications · doi:10.1016/j.eswa.2026.133380 · Cited by 1
IRAISE 2026: Impactful and Responsible AI Systems for Education
Ananda, M., Mallick, D. B., Burstein, J. C., Murphy, A., Wang, Z., & Woodhead, S. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-29794-5_22
An LLM-based multi-agent system for patient discharge assessment
Gebreab, S. A., Musamih, A., & Salah, K. (2026) · Computer Methods and Programs in Biomedicine Update · doi:10.1016/j.cmpbup.2026.100261
Thomas, C. D., Busch, A. H., & Carlo, S. M. (2025) · Journal of Education Finance and Law · doi:10.5406/30678560.50.3.08
“Because of their architecture, GenAI has been known to “hallucinate,” generating text that is not accurate but may still be believable because all outputs are based on real textual patterns (Ji et al., 2023).”
Peng, X., Rao, G.-Z., Zhang, J.-H., & Zhang, L. (2026) · Expert systems with applications · doi:10.1016/j.eswa.2026.133433
Dzialo, J. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-29918-5_8
“Hallucination [10] becomes difficult to diagnose when there is no trace of what the model actually saw, with no steps to reproduce it.”
Building Evidence-Linked Curriculum Knowledge Graphs for Academic Pathway Planning
Zhao, J., Nelson, R. B., Cruz, D. P., et al. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-29744-0_32
From Assessment to Tutoring: Evidence-Led Faithfulness for Goal-Oriented Singing Practice
Cheng, T., & Liu, Z. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-29794-5_70
Hong, D., Kim, Y. J., & Hmelo‐Silver, C. E. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-29794-5_67
Gilson, S., Tabarsi, B. T., & Barnes, T. M. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-29770-9_48
Ghaluh, B. M. (2026) · International Journal of Early Childhood · doi:10.1007/s13158-026-00530-z · Cited by 3
Yıldırım, E., Tunalı, E., & Karaçay, Ş. (2026) · BMC Oral Health · doi:10.1186/s12903-026-09058-z · Cited by 1
“Across all scenarios, unsafe recommendations were rare, contrasting with concerns in the literature regarding hallucinations and potentially harmful outputs [9, 10, 12].”
Simulating Validity: Modal Decoupling in MLLM Generated Feedback on Science Drawings
Bewersdorff, A., Yuruk, N., & Zhai, X. (2026) · International Conference on Artificial Intelligence in Education · doi:10.1007/978-3-032-29760-0_37
“This points away from “generate then justify” [8] and toward “generate only what can be grounded,” with explicit support for verification and abstention when grounding is uncertain.”
Ghorbani, M., Suleimenova, D., Harbach, L., Mazumdar, P., Paramanathan, N., Severino, R., Aydemir, O., & Groen, D. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-29915-4_42
“However, retrieval cannot compensate for stale or malformed real-time context, since retrieval-based grounding remains bounded by the quality and structure of the underlying corpus [6,10].”
Bai, X., Li, P. P., Li, C., & Wu, Y. N. (2026) · International Symposium on Computer Technology and Information Science (ISCTIS) · doi:10.1109/isctis70043.2026.11572187
“LLMs can support context-sensitive extraction and structured output generation, while retrieval modules can narrow the evidence scope and improve contextual relevance [8] .”
FASST-LLM: Implementing Service Scanning Tools Automatically with LLM
Qin, L., Luo, Y., Li, C., Wang, Z., & Yang, J. (2026) · IEEE Conference on Computer Communications (INFOCOM) · doi:10.1109/infocom59046.2026.11571366
“However, because its generation is not grounded in RFC documentation, ChatAFL is susceptible to LLM hallucinations [33].”
Academic Publishing in the Age of Generative AI
Hanneke, B., Hering, F., Pfeiffer, J., Reijers, H. A., & Hinz, O. (2026) · Business & Information Systems Engineering · doi:10.1007/s12599-026-01013-0 · Cited by 3
“There is a risk that these errors compound across studies, systematically degrading research quality (Gopal et al. 2025; Ji et al. 2023).”
Mora-Ruano, D., & Mora-Ruano, R. (2026) · Ensayos: Revista de la Facultad de Educación de Albacete · doi:10.18239/ensayos.v41i2.3949
AIANO: Enhancing Information Retrieval with AI-Augmented Annotation
Khattab, S., Bauer, M., Heine, L., Rostalski, T., Kleesiek, J., & Friedrich, J. (2026) · Lecture notes in computer science · doi:10.1007/978-981-92-1926-1_18
Zhang, J., Tang, J., Li, S., Wang, K., & Wang, T. (2026) · Communications in computer and information science · doi:10.1007/978-981-95-8525-0_12
Enhancing Digital Collaboration Through Human-Centered AI and Machine Learning
Kumar, K. G., Sangamithra, T., Balamurugan, K., Periyasamy, S., Keerthana, N. V., & Vijay, V. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-8217-3.ch004
Can ChatGPT graduate as a civil engineer? Exploring the role of prompt design in model behaviour
Genç, O., Durgun, S., & Un, B. (2026) · Engineering Construction & Architectural Management · doi:10.1108/ecam-01-2026-0023 · Cited by 2
Bai, M., Zhu, M., Zhu, H., Yao, D., Liu, Y., & Ju, J.-W. W. (2026) · Advanced Engineering Informatics · doi:10.1016/j.aei.2026.105017
LLM-Assisted Multi-domain Translation with Composed In-Domain Experts
Shi, X., Zhou, A., Chen, Y., Huang, K., Han, W., & Xu, J. (2026) · Communications in computer and information science · doi:10.1007/978-981-92-0199-0_5
Patel, R. K., Sharma, Y., Rama, J., & Patel, N. R. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-9725-2.ch011
Beyond speed and professionalism: Evidence from a discrete choice experiment on AI coach preferences
Sun, Y., & Hou, Z. (2026) · Journal of Retailing and Consumer Services · doi:10.1016/j.jretconser.2026.104960 · Cited by 1
Moya-Sánchez, D. N. (2026) · AI in neuroscience. · doi:10.1177/2997979x261465520
Zhang, H., Wang, X., Dengxiong, X., Robertson, J., Zhang, Y., Schwartz, D. I., Gao, S. Z., Yuan, X., & Peng, C. (2026) · International Journal of Disaster Risk Reduction · doi:10.1016/j.ijdrr.2026.106269
Agentic Self-RAG: Multi-Agent Reasoning for Self-Correcting Retrieval-Augmented Generation
Hassan, S. B., -, A., & Abbas, M. (2026) · International Conference on IT and Industrial Technologies (ICIT) · doi:10.1109/icit68548.2026.11577708 · Cited by 1
“Researchers refer to this problem as hallucination [2].”
Song, Y., Qiao, S., He, Q., Liu, Z., Li, C., Tian, Z., & Liu, X. (2026) · Advanced Engineering Informatics · doi:10.1016/j.aei.2026.105001
Ballell, T. R. D. L. H. (2025) · Springer eBooks · doi:10.1007/978-3-032-19720-7_13
Dini Bilgi Edinme Sürecinde Yapay Zekâ Tabanlı Sohbet Robotlarının Kullanımına İlişkin Tutum Ölçeği
BAŞTÜRK, H., & Oruç, C. (2026) · Hitit İlahiyat Dergisi · doi:10.14395/hid.1856515
Validating LLM-Generated SQL Queries through Metamorphic Prompting
Lin, L. G., Zhu, Q., Hong, J., Wang, C., Liu, Y., & Wu, R. (2026) · Proceedings of the ACM on software engineering. · doi:10.1145/3797146 · Cited by 1
Yang, W., Zhang, H., Jin, H., CHEN, S., Shi, Q., & Chen, G. (2026) · Discover Computing · doi:10.1007/s10791-026-10325-4
“Beyond the challenge of personalization, LLM-based systems exhibit another well-documented limitation: the tendency to generate plausible but incorrect information, often termed “hallucination” [8].”
AI-Powered Service Innovation and Hyper Personalization in Modern Product Ecosystems
Bhardwaj, A. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-5067-7.ch004
Gao, F., Sun, Y., Li, Y., & Gu, J. (2026) · Communications in computer and information science · doi:10.1007/978-981-95-8525-0_10
Li, Z., 刘晓霞, Chen, T., Yang, Y., Liu, X., Lv, Y., Zhao, Z., Li, X., Yin, X., Feng, Z., et al. (2026) · Journal of Medical Systems · doi:10.1007/s10916-026-02433-x
Güder, Ş. C. (2026) · The Turkish Online Journal of Design Art and Communication · doi:10.7456/tojdac.1929108 · Cited by 2
Cao, X., Li, J., Ying, T., & Liu, Z. (2026) · Information Processing & Management · doi:10.1016/j.ipm.2026.105028 · Cited by 3
Information Seeking in Translation: Bibliometric Analysis and Scoping Review
Xu, Y., Wu, L., Lei, V. L. C., & Li, D. (2026) · Investigación Bibliotecológica Archivonomía Bibliotecología e Información · doi:10.22201/iibi.24488321xe.2026.107.59120 · Cited by 2
“Unlike traditional resources, large language models generate plausible outputs that may contain factual inaccuracies or fabricated terminological equivalents, a phenomenon known as “hallucina-tion” (Guerreiro, Voita, and Martins 2023, 1060; Ji et al. 2023, 3-4).”
Self-citationAllaham, M. V. (2026) · Journal of Computer and Education Research · doi:10.18009/jcer.1767405
Mohanraj, P., & Fatima, T. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-30833-7_35
Zhang, C., Wang, M., You, X., & Gu, Z. (2026) · Research in Veterinary Science · doi:10.1016/j.rvsc.2026.106318
Luo, J., Li, S., Wang, M., Cao, X., & Li, H. (2026) · IEEE Transactions on Learning Technologies · doi:10.1109/tlt.2026.3708984
“A major concern is the “AI hallucination,” in which the GenAI model generates content that is factually incorrect, illogical, or misleading[24].”
Bozkurt, H. T., & Bozkurt, A. (2026) · Journal of Midwifery and Health Sciences · doi:10.62425/esbder.1911961
Specializing Language Models for Textual Fuzzing via Reinforcement Learning
Lin, J., Su, L., Li, J., & Qian, C. (2026) · IEEE Symposium on Security and Privacy (S&P) · doi:10.1109/sp63933.2026.00184
“The scale of these codebases exacerbates well-known LLM limitations, including the context window size and hallucination trade-off [8].”
Anonymous, Chen, X., Xiong, Y., Xu, S., Li, Q., & Zhou, S. (2026) · Physical Review Physics Education Research · doi:10.1103/9t5b-twsb
PromptFishing: Active Hallucination Inducement to Distinguish LLMs From Humans
Chen, B., Li, G., Wu, J., Li, J., & Fang, H. (2026) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2026.3709099
“LLMs are prone to hallucinations, i.e., confidently generating factually incorrect or nonsensical content despite high linguistic fluency [28], [29].”
AUTOMED: A Finite Automata-Based Framework for Health Misinformation Detection in Medical Chatbots
Postiglione, A., & Nota, G. (2026) · Lecture notes on data engineering and communications technologies · doi:10.1007/978-3-032-23347-9_30
Altunkaya, M., Babür, E., Cihan, E., & Pirinççi, C. Ş. (2026) · Work · doi:10.1177/10519815261462135
Narayankeri, S., Sapar, N., Buchade, A., Taralekar, S. S., & Nalavade, J. E. (2026) · International Conference on Computing, Communication and Green Engineering (CCGE) · doi:10.1109/ccge67142.2026.11581753
“Ji et al. found LLMs display an excessive level of certainty about their predictions [9], [10].”
Ahmad, A., Karaky, H., & Shahine, M. Y. A. (2026) · Computer Applications in Engineering Education · doi:10.1002/cae.70227
“While GenAI provides unprecedented speed and scale, it is not an infallible oracle [38].”
Jeong, C.-W., & Jung, H. (2026) · The Journal of Korean Institute of Information Technology · doi:10.14801/jkiit.2026.24.6.33
Üretken Yapay Zekâ ve İktisat Eğitimi: Bizi Ne Bekliyor?
Yolusever, A. (2026) · Iğdır Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi · doi:10.58618/igdiriibf.1946886
Sun, Y., & Sun, J. (2026) · ACM Interactive Health Conference · doi:10.1145/3786579.3804942
Ukrainian RAG System with Morphological Normalization and Fact Verification
Denysiuk, O. (2026) · Artificial Intelligence · doi:10.15407/jai2026.02.036
Evaluating large language models for technology-oriented searches in engineering design
Koutsouvelis, V., Silipigni, F., & Becattini, N. (2026) · Proceedings of the Design Society · doi:10.1017/pds.2026.10597 · Cited by 1
“However, the utility of LLMs is constrained by significant limitations, primarily “ hallucination ” , the generation of plausible but incorrect information (Ji et al., 2023), and a dependence on prompt quality and source credibility (Bsharat et al., 2024).”
Kül, S., Arslan, B., & Çeltek, S. A. (2026) · Global Power, Energy and Communication Conference · doi:10.1109/gpecom70462.2026.11578628
“…the consequence severity of a hallucinated (confident but incorrect) LLM output, a well-documented fail-2026 8th Global Power, Energy and Communication Conference (IEEE GPECOM 2026), June 3-5, 2026, Naples/Italy ure mode of generative language models [12], reaching the actuation or decision layer.”
Temporal Requirement Formalization Using Large Language Models
Hu, Z., Ma, W., Wang, Q., Jiang, L., & Li, D. (2026) · IEEE International Conference on Software Analysis, Evolution and Reengineering (SANER) · doi:10.1109/saner67736.2026.00099
Algorithmic Bias vs. Human Bias: Dual Reflections on Social Work in the AI Era
Liu, Y.-H., Chen, Y.-H., & Liu, J. S.-C. (2026) · Smart innovation, systems and technologies · doi:10.1007/978-3-032-23104-8_16
The Hawash Method: A Bedside-First Framework for Safe Clinical GenAI Decision Support
Hawash, A. M. (2026) · ICEENG International Conference for Innovations in Intelligent Computing and Cybersecurity (IICC) · doi:10.1109/iicc69623.2026.11582187
“Hallucination remains a fundamental limitation [5], [6], with reported frequencies varying substantially across tasks and models [6], [7].”
Nguyen, M. T. H., Vuong, A., Tran, V., & Bui, T. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-30497-1_7
Scalise, P., Bray, N., Ho, R., et al. (2026) · International Wireless Communications and Mobile Computing Conference (IWCMC) · doi:10.1109/iwcmc69287.2026.11579894
“Despite their utility, transformer-based MLLMs exhibit well-documented limitations in high-precision tasks, including hallucination, degraded reliability near context window limits, and inaccurate extraction from complex diagrams [3].”
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.




