Survey of Hallucination in Natural Language Generation
Open accessThe DOI matches its Crossref record; 355 of the 360 references it cites were found in a registry or an index, and 4 are not checked yet; 224 of the 256 sources its claims cite were checked for support.
Abstract
Natural Language Generation (NLG) has improved exponentially in recent years thanks to the development of sequence-to-sequence deep learning technologies such as Transformer-based language models. This advancement has led to more fluent and coherent NLG, leading to improved development in downstream tasks such as abstractive summarization, dialogue generation, and data-to-text generation. However, it is also apparent that deep learning based generation is prone to hallucinate unintended text, which degrades the system performance and fails to meet user expectations in many real-world scenarios. To address this issue, many studies have been presented in measuring and mitigating hallucinated texts, but these have never been reviewed in a comprehensive manner before. In this survey, we thus provide a broad overview of the research progress and challenges in the hallucination problem in NLG. The survey is organized into two parts: (1) a general overview of metrics, mitigation methods, and future directions, and (2) an overview of task-specific research progress on hallucinations in the following downstream tasks, namely abstractive summarization, dialogue generation, generative question answering, data-to-text generation, and machine translation. This survey serves to facilitate collaborative efforts among researchers in tackling the challenge of hallucinated texts in NLG.
Verified 2026-11-19 · 4 references not checked yet
Overview
CoChat Summary
TL;DR
This survey provides a broad overview of hallucination in natural language generation, covering definitions, contributors, metrics, mitigation methods, and task-specific progress across summarization, dialogue, GQA, data-to-text, machine translation, vision-language generation, and LLMs. It distinguishes intrinsic from extrinsic hallucination and argues that hallucination must be treated differently across tasks and sources.
Key findings
- The survey categorizes hallucination into intrinsic hallucination, where generated output contradicts the source, and extrinsic hallucination, where output cannot be verified from the source; extrinsic hallucination is not always erroneous because it may come from factually correct external information.
- It reports prevalence examples from cited work: 25% of summaries generated by state-of-the-art abstractive summarization systems have hallucinated content; 62% of first sentences in WIKIBIO have additional information not stated in the corresponding infobox; and 74% of samples in the QMSum dataset consist of inconsistent facts.
- It reviews hallucination metrics as statistical, model-based (IE-based, QA-based, NLI-based, faithfulness classification, LM-based), and human evaluation, and mitigation methods as data-related and modeling/inference methods including architecture changes, planning/sketching, RL, multi-task learning, controllable generation, and post-processing.
- For LLMs, it notes that RLHF reduced the hallucination rate from 41% to 21% in closed-domain tasks for InstructGPT, and discusses reference-dependent and reference-free hallucination metrics plus mitigation via data refinement, RAG, safety fine-tuning, model editing, decoding methods, chain-of-thought variants, post-processing, and ensembling.
Methods
- Literature survey design: the paper reviews existing research on hallucination in NLG rather than conducting new experiments or primary data collection.
- Coverage: general NLG hallucination topics (definitions, contributors, metrics, mitigation, future directions) plus task-specific sections on abstractive summarization, dialogue generation, generative question answering, data-to-text generation, neural machine translation, vision-language generation, and large language models.
- Analysis: synthesizes and compares definitions, metrics, mitigation approaches, and future directions across tasks, with examples in tables and figures; the LLM section was updated in Jan 2024.
Limitations
- The paper states that GQA has no standard definition or categorization of hallucination and no automatic metric specifically for hallucination in GQA; it also notes that open-domain dialogue lacks a standard hallucination metric and that VL hallucination research is still in its early stage.
- The paper states that extrinsic hallucination has been somewhat overlooked and is more challenging to reduce, and that most existing mitigation methods do not distinguish between intrinsic and extrinsic hallucination.
- Evident from the text: as a survey, it does not provide new empirical validation of metrics or mitigation methods, and its LLM coverage is explicitly updated only through Jan 2024, so later work is not included.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures
Verification
Claims the paper makes
124 Valid · 26 Not valid · 19 Not checked · 135 Own results and interpretations
At least one cited source doesn’t support this claim or points to a different paper, and none confirmed it — see the verdicts below.
In the paper
“Furthermore, perturbation, where the insertions of most common tokens are placed at the beginning of the sentence, seems to be the most successful in hallucination mitigation.”
What the sources say
- Not checked — no full text of the sourcecites [133] [133] Katherine Lee, Orhan Firat, Ashish Agarwal, Clara Fannjiang, and David Sussillo. 2019. Hallucinations in Neural Machine Translation. ICLR (2019). -
- Disputedcites [215] [215] Vikas Raunak, Arul Menezes, and Marcin Junczys-Dowmunt. 2021. The Curious Case of Hallucinations in Neural Machine Translation. (4 2021), 1172-1183. https://arxiv.org/abs/2104.06683v1 - sourceExperimental — accuracy not yet measured
In the passage describing the perturbation procedure (page unavailable), the paper says that inserting common tokens at the beginning is used to generate hallucinations, not mitigate them. Its prevention section instead discusses approaches such as robust learning and corpus-level filtering. (LLM, conf=0.92)
Checked by a reasoning model — may be wrong.
hedgedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
At least one cited source doesn’t support this claim or points to a different paper, and none confirmed it — see the verdicts below.
In the paper
“While there are works on the visual extension of faithfulness-oriented RLHF [303], these essentially involve labeling more data, which is expensive and difficult to scale.”
What the sources say
- Disputedcites [303] [303] Tianyu Yu, Yuan Yao, Haoye Zhang, Taiwen He, Yifeng Han, Ganqu Cui, Jinyi Hu, Zhiyuan Liu, Hai-Tao Zheng, Maosong Sun, and Tat-Seng Chua. 2023. RLHF-V: Towards Trustworthy MLLMs via Behavior Alignment from Fine-grained Correctional Human Feedback. CoRR abs/2312.00849 (2023). https://doi.org/10.48550/ARXIV.2312.00849 arXiv:2312.00849 - sourceExperimental — accuracy not yet measured
RLHF-V does use human annotation for visual-model alignment, but it makes the contrary point about needing more labeled data. The Fine-grained Correctional Human Preference Collection section says its corrections improve data efficiency, and Section 4.3 reports comparable hallucination rates using only 200 preference examples versus an order of magnitude more ranking-labeled examples. Our copy does not establish a separate claim about annotation cost or scalability. (LLM, conf=0.88)
Checked by a reasoning model — may be wrong.
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Disputed
Every source the paper cites for this claim was checked and supports it.
In the paper
“Nevertheless, as illustrated by Falke et al. [66], off-the-shelf NLI models tend to transfer poorly to the abstractive summarization task.”
What the sources say
- Supportedcites [66] [66] Tobias Falke, Leonardo FR Ribeiro, Prasetya Ajie Utama, Ido Dagan, and Iryna Gurevych. 2019. Ranking Generated Summaries by Correctness: An Interesting but Challenging Application for Natural Language Inference. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics . 2214-2220. - source
The Introduction states that models trained on NLI datasets transfer poorly to the paper’s downstream task of ranking generated summaries by correctness. The Conclusions likewise say that out-of-the-box NLI models do not perform well on that task. (LLM, conf=0.94)
- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Such wrong association results in generation with facts mixed up between two similar entities [60, 233].”
What the sources say
- Partly supportedcites [60] [60] Nouha Dziri, Andrea Madotto, Osmar Zaiane, and Avishek Joey Bose. 2021. Neural Path Hunter: Reducing Hallucination in Dialogue Systems via Path Grounding. EMNLP (2021). - sourceExperimental — accuracy not yet measured
The GPT2-KG passage describes an incorrect association between entities, and G Error Analysis gives an example of a generated response assigning a book to the wrong author. These support the claim that facts can be mixed up, but the paper does not say the entities involved are similar. The supplied passages do not provide page numbers. (LLM, conf=0.83)
- Supportedcites [233] [233] Kurt Shuster, Spencer Poff, Moya Chen, Douwe Kiela, and Jason Weston. 2021. Retrieval Augmentation Reduces Hallucination in Conversation. EMNLP (2021). - source
The Introduction states that language models often mix up facts between two similar entities when generating factually incorrect statements. This directly supports the cited description of the resulting error. (LLM, conf=0.92)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“As a result, the target reference may contain information that cannot be supported by the source [198, 268].”
What the sources say
- Supportedcites [198] [198] Ankur Parikh, Xuezhi Wang, Sebastian Gehrmann, Manaal Faruqui, Bhuwan Dhingra, Diyi Yang, and Dipanjan Das. 2020. ToTTo: A Controlled Table-To-Text Generation Dataset. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) . 1173-1186. - source
The Dataset section says that reference targets selected from web sentences may contain information that cannot be inferred from the source. The Data Annotation Process section also describes deleting phrases unsupported by table cells. (LLM, conf=0.93)
- Supportedcites [268] [268] Hongmin Wang. 2019. Revisiting Challenges in Data-to-Text Generation with Fact Grounding. In Proceedings of the 12th International Conference on Natural Language Generation . 311-322. - source
The Introduction reports that about 40% of human-written game-summary content cannot be directly mapped to the input boxscore records and draws on information beyond those inputs. Section 2.1 likewise describes reference content not licensed by the boxscore table. This supports the claim that a target reference may contain information the source does not support. (LLM, conf=0.93)
hedgedstates a possible cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
117 more claims · 74 with their cited sources checked
Create a free account to see all 304 claims and to verify the 2 references still open.
- Borderline: one of the tests for a substantial claim is unclear. 1
- It may state two separate claims as one. 90
- Probably not a substantial claim. 142
- The quote only partly states this claim. 1
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Hallucination can lead to serious consequences such as incorrect or misleading code, security risks, and ethical issues [186].”
From A survey on large language model based autonomous agents (2024) · cites this paper as [186]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“citation fabrication constitutes a form of extrinsic hallucination in the absence of retrieval augmentation, since the model generates references not grounded in any provided source material [4].”
From How LLMs Cite and Why It Matters: A Cross-Model Audit of Reference Fabrication in AI-Assisted Academic Writing and Methods to Detect Phantom Citations (2026) · cites this paper as [4]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Recent taxonomic work distinguishes at least seven cases [15,18].”
From Hallucination as epistemic pollution: epistemic injustice and credibility in futures-oriented science education (2026) · cites this paper as [18]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“LLMs struggle to generate accurate and up-to-date responses on current topics, due to outdated knowledge, lack of domain-specific expertise, or hallucination13–17.”
From Leveraging long context in retrieval augmented language models for medical question answering (2025) · cites this paper as [16]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Fine-tuning can enrich LLMs with domain knowledge, enhance their specific ability, improve the fairness and reliability of the outputs, and prevent certain damage caused by hallucination [26].”
From FederatedScope-LLM: A Comprehensive Package for Fine-tuning Large Language Models in Federated Learning (2024) · cites this paper as [26]
Experimental — accuracy not yet measured
41 more claims about this paper
Create a free account to see every claim other papers make about this one.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Extrinsic Object Hallucination
“CHAIR (Caption Hallucination Assessment with Image Relevance) metric, which calculates what proportion of object words generated are actually in the image according to the ground truth captions.”
§ Extrinsic Object Hallucination
“CHAIR 𝑖 measures per-instance object hallucination, i.e. what fraction of object instances in each generated caption are hallucinated.”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ 7.1 Hallucination Definition in Abstractive Summarization
“For instance, in Table 1, given the input article shown in the caption, an example of intrinsic hallucination is ' The Ebola vaccine was rejected by the FDA in 2019, '”
§ 2.1 Categorization
“in the abstractive summarization task from Table 1, the generated summary ' The first Ebola vaccine was approved in 2021 ' contradicts the source content”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Survey of Hallucination in Natural Language Generation
“The survey is organized into two parts: (1) a general overview of metrics, mitigation methods, and future directions; (2) an overview of task-specific research progress on hallucinations in the following downstream tasks, namely abstractive summarization, dialogue generation, generative question answering, data-to-text generation, machine translation, and visual-language generation; and (3) hallucinations in large language models (LLMs)”
§ 1 INTRODUCTION
“The third part discusses this phenomenon in LLMs in Section 13.”
Citations
Cites
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325 · cited 3×
“Recently, the rapid development of NLG has captured the imagination of many thanks to the advances in deep learning technologies, especially Transformer [189]-based models like BERT [29], BART [100], GPT-2 [149], and GPT-3 [16].”
BackgroundAn Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., et al. (2020) · arXiv · doi:10.48550/arxiv.2010.11929 · Cited by 47,472
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
Sequence to Sequence Learning with Neural Networks
Sutskever, I., Vinyals, O., & Le, Q. V. (2014) · arXiv · doi:10.48550/arxiv.1409.3215 · Cited by 17,766
Advances in Neural Information Processing Systems 29
Teymur, O., Zygalakis, K., & Calderhead, B. (2016) · Cited by 13,887
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., Zhou, Y., Li, W., & Liu, P. J. (2019) · arXiv · doi:10.48550/arxiv.1910.10683 · Cited by 10,929 · cited 1×
“This work investigates the performances of GPT-3 [13], GPT-Neo/J [141], GPT-2 [105] and a T5-based model [106].”
BackgroundAdvances in Neural Information Processing Systems 14: Proceedings of the 2001 Conference
(2002) · The MIT Press eBooks · doi:10.7551/mitpress/1120.001.0001 · Cited by 8,890
Lewis, M., Liu, Y., Goyal, N., Ghazvininejad, M., Mohamed, A., Levy, O., Stoyanov, V., & Zettlemoyer, L. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.703 · Cited by 8,821 · cited 1×
“Recently, the development of NLG has conquered the imagination of many thanks to the advances in deep learning technologies, especially the Transformer [159]-based models like BERT [24], BART [83], GPT-2 [127], and GPT3 [13].”
BackgroundLewis, M., Liu, Y., Goyal, N., Ghazvininejad, M., Mohamed, A., Levy, O., Stoyanov, V., & Zettlemoyer, L. (2019) · arXiv · doi:10.48550/arxiv.1910.13461 · Cited by 8,816 · cited 1×
“Recently, the development of NLG has conquered the imagination of many thanks to the advances in deep learning technologies, especially the Transformer [159]-based models like BERT [24], BART [83], GPT-2 [127], and GPT3 [13].”
Background
Research works printed but not linked: 16
- [4] Joshua Albrecht and Rebecca Hwa. 2007. A Re-examination of Machine Learning Approaches for Sentence-Level MT Evaluation. In Proceedings of the 45th Annual Meeting of the Association of Computational Linguistics . Association for Computational Linguistics, Prague, Czech Republic, 880-887. https://aclanthology.org/P07-1111 -
- [16] Andrzej Białecki, Robert Muir, Grant Ingersoll, and Lucid Imagination. 2012. Apache lucene 4. In SIGIR 2012 workshop on open source information retrieval . 17. -
- [53] Yilun Du, Shuang Li, Antonio Torralba, Joshua B Tenenbaum, and Igor Mordatch. 2023. Improving Factuality and Reasoning in Language Models through Multiagent Debate. In Forty-first International Conference on Machine Learning . -
- [121] Philipp Koehn and Rebecca Knowles. 2017. Six Challenges for Neural Machine Translation. In First Workshop on Neural Machine Translation . Association for Computational Linguistics, 28-39. -
- [145] Junnan Li, Dongxu Li, Caiming Xiong, and Steven Hoi. 2022. BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation. In ICML . -
- [149] Tian Li, Ahmad Beirami, Maziar Sanjabi, and Virginia Smith. 2020. Tilted Empirical Risk Minimization. In International Conference on Learning Representations . -
- [164] Jiasen Lu, Jianwei Yang, Dhruv Batra, and Devi Parikh. 2018. Neural Baby Talk. In CVPR . -
- [165] Yukun Ma, Khanh Linh Nguyen, Frank Z Xing, and Erik Cambria. 2020. A Survey on Empathetic Dialogue Systems. Information Fusion 64 (2020), 50-70. -
- [175] Marianna J. Martindale, Marine Carpuat, Kevin Duh, and Paul McNamee. 2019. Identifying Fluently Inadequate Output in Neural and Statistical Machine Translation. In MTSummit . -
- [179] Kathleen McKeown. 1992. Text Generation . Cambridge University Press. -
- [209] Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, and Ilya Sutskever. 2019. Language Models are Unsupervised Multitask Learners. OpenAI Blog 1, 8 (2019), 9. -
- [238] Hui Su, Xiaoyu Shen, Sanqiang Zhao, Zhou Xiao, Pengwei Hu, Cheng Niu, and Jie Zhou. 2020. Diversifying Dialogue Generation with Non-Conversational Text. In 58th Annual Meeting of the Association for Computational Linguistics . ACL, 7087-7097. -
- [256] Zhaopeng Tu, Yang Liu, Lifeng Shang, Xiaohua Liu, and Hang Li. 2017. Neural Machine Translation with Reconstruction. In Thirty-First AAAI Conference on Artificial Intelligence . -
- [261] Neeraj Varshney, Wenlin Yao, Hongming Zhang, Jianshu Chen, and Dong Yu. 2023. A Stitch in Time Saves Nine: Detecting and Mitigating Hallucinations of LLMs by Validating Low-Confidence Generation. CoRR abs/2307.03987 (2023). https://doi.org/10.48550/arXiv.2307.03987 arXiv:2307.03987 - · arXiv · doi:10.48550/arxiv.2307.03987 · Paper page
- [285] Zeqiu Wu, Michel Galley, Chris Brockett, Yizhe Zhang, Xiang Gao, Chris Quirk, Rik Koncel-Kedziorski, Jianfeng Gao, Hannaneh Hajishirzi, Mari Ostendorf, et al. 2021. A Controllable Model of Grounded Response Generation. In Proceedings of the AAAI Conference on Artificial Intelligence , Vol. 35. 14085-14093. -
- [291] Xinnuo Xu, Ondřej Dušek, Verena Rieser, and Ioannis Konstas. 2021. AGGGEN: Ordering and Aggregating while Generating. roceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL2021) (2021). -
Cited by
Items 2,901–3,000 of 4,844
Parmar, D., Shah, D., Mazumdar, P., & Mallik, S. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5355218
Borhani, S., & Jiang, X. (2025) · medRxiv · doi:10.1101/2025.08.05.25332968
Khatiwada, H. (2025) · Research Square · doi:10.21203/rs.3.rs-7338701/v1
Understanding Generative AI from Multidisciplinary Perspectives
Zou, M., Poncibò, C., Ebers, M., & Calo, R. (2025) · Cambridge University Press eBooks · doi:10.1017/9781009492553.003
VIE-Net: Regressive U-Net for Vegetation Index Estimation
Capparella, V., Nemmi, E. N., Violino, S., Costa, C., Figorilli, S., Moscovini, L., Pallottino, F., Pane, C., Mei, A., & Ortenzi, L. (2025) · IEEE Access · doi:10.1109/access.2025.3598124 · Cited by 3
“In practice, generative neural networks, instead of giving no answer, distort, alter, or create realities of various kinds and are not always reliable [13], [14], [15].”
A Path Toward Collective Superintelligence: AGI, CMI and Human-AI CollaborationPreprint
Wedel, K. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5373169
OpGPT: LLM-driven Operator Generation Prompt Framework for Domain-Specific Architectures
Wu, Y., Dou, Q., Zhao, G., Yuan, J., Chen, R., & Wang, L. (2025) · Procedia Computer Science · doi:10.1016/j.procs.2025.07.117
Multi-Agent Collaboration for Hallucination Correction in Complaint Management
ÇELDİR, O. M., & Dalkılıç, G. (2025) · International Symposium on Innovative Approaches in Smart Technologies (ISAS) · doi:10.1109/isas66241.2025.11101785
“Hallucinations are divided into two main groups: extrinctive and instrictive [4].”
Liu, Q., Wang, Y., Mu, L., & Li, J. (2025) · Discover Computing · doi:10.1007/s10791-025-09675-2
Design of Multimodal RAG Architecture for Intelligent Mixed Reality
Kim, H., Bae, J., Jeong, W., & Park, S. (2025) · Journal of the Korea Computer Graphics Society · doi:10.15701/kcgs.2025.31.3.57
Singh, A., Ghosh, S. K., Singh, K., & Lepri, B. (2025) · IEEE Transactions on Computational Social Systems · doi:10.1109/tcss.2025.3587696
“Models also face hallucination, generating irrelevant content not grounded in inputs, especially in visually detailed tasks [23], [24].”
A General Model for Representing Knowledge and Its Application to Intelligent Systems Based on AI
Fehlmann, T., & Kranich, E. (2025) · Lecture notes in networks and systems · doi:10.1007/978-3-031-92602-0_33
MELLM - An Automatic Evaluation Framework for LLM Without Human Supervision
Cui, C., Zhu, L., Ning, H., Sheng, X., & Wu, H. (2024) · International Conference on Big Data, Artificial Intelligence and Internet of Things Engineering (ICBAIE) · doi:10.1109/icbaie63306.2024.11116990
“But for human questions with complex logical reasoning, factual judgment and structured data output, this method still remains uncertain [13].”
GenAI in Patent DraftingPreprint
Xie, W. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5392009
Insights into Faculty's use of Generative Artificial Intelligence systems in Engineering Classrooms.
Hassani, A., Daher, T. A., Trainin, G., & Wheeler, J. M. (2025) · ASEE Annual Conference & Exposition · doi:10.18260/1-2--56806 · Cited by 2
Tu, J., & Tang, M. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-02534-0_11 · Cited by 1
Foundations of LLMs and Financial Applications
Chung, Y., Kim, J., Kim, M., Joo, M., & Cho, H. (2025) · Blockchain technologies · doi:10.1007/978-981-96-5833-6_4 · Cited by 2
Differential diagnosis making with large language models and probabilistic logic program
Galitsky, B. A. (2025) · Elsevier eBooks · doi:10.1016/b978-0-443-30046-2.00010-7
Bogadi, C. R. (2025) · International Journal of Computer Trends and Technology · doi:10.14445/22312803/ijctt-v73i7p101 · Cited by 1
FTC probe of OpenAI: Consumer protection is the opening salvo of US AI regulation
Susarla, A. (2023) · The Conversation · doi:10.64628/aai.5pa4kwcw3
Yu, H. Q., Scanlon, B., & Reiff‐Marganiec, S. (2025) · IEEE International Symposium on Service-Oriented System Engineering (SOSE) · doi:10.1109/sose67019.2025.00005
Giacalone, M. (2025) · Research Square · doi:10.21203/rs.3.rs-7446174/v1
Informing AI Prompt Engineering for Flightcrew Training Design Through Human Factors Research
Duruaku, F., Sonnenfeld, N. A., Nguyen, B., & Jentsch, F. G. (2025) · Proceedings of the Human Factors and Ergonomics Society Annual Meeting · doi:10.1177/10711813251369798 · Cited by 1
“( Alkaissi & McFarlane, 2023; Ji et al., 2023).”
AI-fueled deception: mechanics of machine-driven manipulation
Shin, D. (2025) · Elsevier eBooks · doi:10.1016/b978-0-443-16104-9.00006-3
Santos, G. R., Fleury, A. L., Leite, R., Paz, J. C. R. A., Ortega, F. J. M., Mamillapalli, L. S., Luduvico, D., Soares, M., Murakami, M., & Zancul, E. D. S. (2025) · Procedia CIRP · doi:10.1016/j.procir.2025.08.103 · Cited by 2
Reduction of supervision for biomedical knowledge discovery
Theodoropoulos, C., Coman, A. C., Henderson, J., & Moens, M. (2025) · BMC Bioinformatics · doi:10.1186/s12859-025-06187-0 · Cited by 1
“While using LLMs in zero-shot or few-shot learning setups can reduce the need for supervision, challenges remain regarding the models’ stability and hallucinations [45, 46, 47, 48].”
Heo, S., Choi, W., & Lee, S. (2025) · Journal of the Computational Structural Engineering Institute of Korea · doi:10.7734/coseik.2025.38.4.247
Heo, S., Choi, W., & Lee, S. (2025) · Journal of the Computational Structural Engineering Institute of Korea · doi:10.7734/coseik.2025.38.4.257
Knowledge Intensive AgentsPreprint
Liu, Z., Liu, Z., Huang, P., et al. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5459034
Cheng, W., Guo, J., & Lu, D. (2025) · Disability and Rehabilitation Assistive Technology · doi:10.1080/17483107.2025.2554955 · Cited by 2
““highly effective” was defined as ratings of 4 or 5 on a 5-point scale (1=“Not effective at all”, 5 = extremely effective”). at the same time, the researchers we worked with raised two key pedagogical concerns: whether error reduction reflects true learning versus mechanical correction, and whether…”
RetractedNavarro, D. S. (2025) · Revista Científica de la Facultad de Derecho y Ciencias Sociales y Políticas · doi:10.30972/rcd.428685
Liu, Y. (2025) · Lecture notes in networks and systems · doi:10.1007/978-3-031-94892-3_28
Balanced KGQA Through Prompt-Based Data Augmentation with LLMs
Ren, X., & Wang, Z. (2025) · Lecture notes in networks and systems · doi:10.1007/978-3-031-94892-3_33
Yuan, X.-B., & Han, R. (2025) · International Conference on Computer and Multimedia Technology (ICCMT) · doi:10.1145/3757749.3757836
A Unified Framework for Knowledge-Intensive Numerical Reasoning over Financial Document
Yin, L., Yin, K., & Zhao, H. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-04627-7_3
How Close Is ChatGPT to Developer Judgment? A Study on Stack Overflow Java Questions
Reis, F. A. G., Maia, M. D. A., & Dantas, C. E. C. (2025) · doi:10.5753/cbsoft_estendido.2025.14164
“Another challenge is the phenomenon of hallucinations [Ji et al. 2023], in which the model generates content that is non-existent or not recognizable to human observers [Zuccon et al. 2023], including code or references that appear plausible but lack reliable provenance [Bifolco et al. 2025].”
Deceit and Power: Machine Learning and Misalignment
Ngo, R., & Bales, A. (2025) · Oxford University Press eBooks · doi:10.1093/9780191979972.003.0026 · Cited by 2
Tan, X., Li, B., Xu, W., Qu, C., Chu, W., Xu, Y., Qi, Y., & Qiu, X. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-05127-1_40 · Cited by 6
Self-citationRadke, R., Diller, G., Reddy, R. G., Shivaram, P., Danford, D. A., & Kutty, S. (2025) · European Heart Journal - Digital Health · doi:10.1093/ehjdh/ztaf111 · Cited by 4
“17,18 While current LLMs primarily store knowledge within their parametric memory and are inherently limited by being trained on a dataset containing information up to a certain knowledge cutoff date, our use of RAG integrates external, verifiable knowledge sources to provide up-to-date and…”
Tei, U., Shiraishi, M., Moriwaki, Y., et al. (2025) · Plastic & Reconstructive Surgery · doi:10.1097/prs.0000000000012478
Integrating Generative AI into Clinical Practice
Quinn, C. E. (2025) · Springer eBooks · doi:10.1007/978-3-032-01613-3_11
A Little Help Goes a Long Way: Tutoring LLMs in Solving Competitive Programming Through Hints
Zhang, Y., Dong, W., Liu, J., Wang, S., Wang, D., Ma, T., Li, Y., & Yang, K. (2025) · IEEE Transactions on Software Engineering · doi:10.1109/tse.2025.3614204 · Cited by 4
“Some studies [19], [20] show that grounding LLMs in domain-specific and trustworthy data can substantially reduce hallucinations and improve output reliability, which aligns with our approach.”
Responsible AI in Action: Navigating the Ethical Landscape of ChatGPT
Tiwari, A., & Kumar, P. (2025) · IGI Global eBooks · doi:10.4018/979-8-3373-0877-7.ch008
ChatGPT as a Catalyst for Innovation: Unlocking Value in Knowledge Economies
Kumar, V. (2025) · IGI Global eBooks · doi:10.4018/979-8-3373-0877-7.ch003
Li, Z., Chen, H., Tan, H., Lan, L., Sui, Y., & Ren, J. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-06078-5_16 · Cited by 1
GLMFuzz: Vulnerability Knowledge Guided Prompting for Efficient Network Protocol Fuzzing
Zhu, S., Gan, S., Wang, X., & Nan, Y. (2025) · IEEE International Conference on Software Quality, Reliability and Security (QRS) · doi:10.1109/qrs65678.2025.00062 · Cited by 2
“However, during the content generation process, LLM often produces information that does not exist, which is a phenomenon known as the ”hallucination” problem [34].”
Wang, Y., Wang, H., Wang, K., & Du, X. (2025) · International Conference on Computer Information Science and Application Technology (CISAT) · doi:10.1109/cisat66811.2025.11181923 · Cited by 1
“We analyze this situation as being caused by two main factors: On the one hand, for SLM, a large number of error cases are the result of faithfulness hallucination [23] .”
Entity hallucination index for text summarization models
Katwe, P., Balabantaray, R. C., & Vittala, K. P. (2025) · Applied Computing and Informatics · doi:10.1108/aci-09-2024-0371 · Cited by 2
Liu, H., Dai, L., & Jiang, H. (2025) · Applied Sciences · doi:10.3390/app151910696
Generative KI in der politikorientierten Technikfolgenabschätzung
Jahnel, J., Nentwich, M., Riousset, P., Milchram, C., Kehl, C., & Albrecht, S. (2025) · Springer eBooks · doi:10.1007/978-3-658-48491-0_9
Reflection, Confabulation, and Reasoning
Nagel, J. (2025) · Wiley eBooks · doi:10.1002/9781394374366.ch6 · Cited by 3
LA INTELIGENCIA ARTIFICIAL GENERATIVA EN LA PLANIFICACIÓN CURRICULAR: ENTRE ALUCINACIONES Y CERTEZAS
Collazos, J. M. S., & Quintero, A. B. (2025) · Dialectica · doi:10.56219/dialctica.v1i26.4414
Extracting information and sentiment analysis on dialogue in financial results briefing
Nakatsuka, H., & Suimon, Y. (2025) · Digital Finance · doi:10.1007/s42521-025-00159-y
“Hallucination affects the reliability of an LLM’s output (Ji et al., 2023).”
Moosa, M., Malumi, O., Chinedu, S., Okah, W., Ogboye, A., Abiakam, C., Onyenemerem, P., Etuk, I., & Mobisson, N. (2025) · doi:10.12688/verixiv.2125.1
No news is good news? The declining information value of broadcast news in America
Gaebler, J., Westwood, S. J., Iyengar, S., & Goel, S. (2025) · PLoS ONE · doi:10.1371/journal.pone.0331607 · Cited by 2
“Unlike human annotators, who often introduce errors when they are insufficiently attentive [46], LLMs are difficult to “distract,” but can hallucinate or exhibit temporal biases [47,48].”
A Sound Static Analysis Approach to I/O API Migration
Li, S., Zhang, Z., Zhong, S., Zhou, D., & Shen, J. (2025) · Proceedings of the ACM on Programming Languages · doi:10.1145/3763071 · Cited by 3
Aligning AI Impact with Strategy: Cross Sector Metrics for Sustainable Business Model Transformation
Meutia, K. I. (2024) · Novatio Journal of Management Technology and Innovation · doi:10.61978/novatio.v2i3.850 · Cited by 1
“Modern AI systems are increasingly autonomous, capable of self-learning, making decisions in ambiguous contexts, and reconfiguring workflows dynamically (Ji et al., 2023).”
Domain-Specific Question Answering System of Large Language Model Enhanced by Knowledge Graph
Liu, S., Chen, B., Li, J., Sun, X., Tang, J., & Zhang, W. (2025) · International Symposium on Machine Learning and Media Computing (MLMC) · doi:10.1109/mlmc65154.2025.11189882
Condor: A Code Discriminator Integrating General Semantics With Code Details
Liang, Q., Zhang, Z., Liu, C., Sun, Z., Zhang, W., Chen, Y., Zhao, Z., Luo, Q., Wang, W., Jiang, Y., et al. (2025) · IEEE Transactions on Software Engineering · doi:10.1109/tse.2025.3620145 · Cited by 6
“However, they typically struggle to produce correct code on the first attempt when addressing complex requirements such as advanced algorithmic problems [7], [8], [9], [10].”
Huang, J., Li, F., Yin, P., et al. (2025) · Chinese Control Conference (CCC) · doi:10.23919/ccc64809.2025.11178926
“Qwen2 is a new chapter in Qwen's large-scale language modeling series, and Qwen2-7B-Instruct increases the number of languages to 30 compared to previous versions, with good Chinese language capabilities and context lengths of up to 128 K [8] .”
Künstliche Intelligenz – eine Schlüsseltechnologie
Hirsch‐Kreinsen, H. (2025) · Fehlzeiten-Report · doi:10.1007/978-3-662-71885-8_1
Ward, R., Dalal, D., Buitelaar, P., & Breslin, J. G. (2025) · ACM SIGAPP Applied Computing Review · doi:10.1145/3770865.3770867 · Cited by 1
A Teacher-Driven Framework for Reliable and Personalised AITutors
Valtolina, S., Matamoros, R. A., & Epifania, F. (2025) · Biannual Conference of the Italian SIGCHI Chapter (CHItaly) · doi:10.1145/3750069.3750121
Fang, X., Yuan, H. R., Li, H., Kou, J., Gu, C., Zhang, W., Duan, X., & Fang, Y. (2024) · International Conference on Universal Village (UV) · doi:10.1109/uv63228.2024.11189217 · Cited by 5
“This consolidation echoes Ji et al.’s [35] manifestation–level taxonomy: Factual-ity, concerning whether the output contradicts ground truth; Faithfulness, concerning whether the output is inconsistent with user-provided context [26], [31], [35].”
De-identification of open-source intelligence using finetuned LLaMA-3
Sun, C., Chen, S.-S., & Ho, Y.-H. (2025) · High-Confidence Computing · doi:10.1016/j.hcc.2025.100357 · Cited by 2
Knowledge prompting: How knowledge engineers use generative AI
Koutsiana, E., Walker, J., Nwachukwu, M., Zhang, B., Meroño-Peñuela, A., & Simperl, E. P. B. (2025) · Journal of Web Semantics · doi:10.1016/j.websem.2025.100873 · Cited by 13
Orellana, S.-M. J., Palacios-Morocho, M. R., & Conde-Zhingre, L. E. (2025) · Communications in computer and information science · doi:10.1007/978-3-032-08366-1_10 · Cited by 3
Holst, C., Wiegräbe, F., Redecker, C., & Lohweg, V. (2025) · Springer eBooks · doi:10.1007/978-3-658-48573-3_11 · Cited by 2
Li, J., & Zhang, P. (2025) · Sustainability · doi:10.3390/su17209051 · Cited by 2
Enhancing Factual Knowledge Retrieval Capabilities of Small Language Models for Natural Hazard
Bautista, L. R. P., Jiménez, S., & Vilches‐Blázquez, L. M. (2025) · Communications in computer and information science · doi:10.1007/978-3-032-06760-9_2
Osmani, Z., & Althonayan, A. (2025) · Communications in computer and information science · doi:10.1007/978-3-032-07370-9_37
Beyond the List: A Framework for the Design of Next-Generation MEDLINE Search Tools
Zhurov, V., Sedig, K., & Milani, M. (2025) · Data · doi:10.3390/data10100167
“Hallucinations are outputs consisting of unfaithful or nonsensical text [66].”
Utilization of Large Language Models for conformity assessment: Chances, Threats, and Mitigations
Litzinger, J., Peters, D., Thiel, F., & Tschorsch, F. (2025) · Annals of Computer Science and Information Systems · doi:10.15439/2025f7772
DR-VQA: Decompose-then-Reconstruct for Visual Question Answering in BLV Assistance
Pan, B., Shi, H., & Gao, X. (2025) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3746027.3755393
AIGC-Enhanced UAV-Based 3D Mapping and Trajectory Planning for Rapid Disaster Response
Zhang, X., Gao, H., Zhang, B., et al. (2025) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3746027.3754708
The Evolution and Latest Trends of AI Technology
Yu, J. (2025) · Research of Chinese education · doi:10.1007/978-981-96-9005-3_1 · Cited by 1
Lu, R., Fan, Z., Wang, G., & Shi, Q. (2025) · Applied Sciences · doi:10.3390/app152111393
A Knowledge Graph Approach Towards Detecting Large Language Model Hallucination
Winter, T. S., & van Vuuren, J. H. (2025) · Lecture notes in networks and systems · doi:10.1007/978-3-032-08384-5_19 · Cited by 1
CPU-Efficient Verification of Science Problem–Solution Pairs: Design Rationale and Baselines
Douard, N., Samet, A., Giakos, G., & Cavallucci, D. (2025) · IFIP advances in information and communication technology · doi:10.1007/978-3-032-08847-5_18
How Faithful Are Your Summaries? A Study of NLI-Based Verification in Portuguese
Paula, F. S. F., Westhelle, M., Corrêa, M. C. M., Bencke, L. R., & Moreira, V. P. (2025) · Simpósio Brasileiro de Tecnologia da Informação e da Linguagem Humana (STIL) · doi:10.5753/stil.2025.37834
“Despite its superior results in comparison to extractive summarization [Feijo and Moreira 2019, El-Kassas et al. 2021, Sharma and Sharma 2022], abstractive summaries may not be faithful to the source text due to hallucinations – a known problem in generative models [Ji et al. 2022].”
A Survey of Large Language Models for Tabular Data Imputation: Tuning Paradigms and Challenges
Jha, S., Goyal, V., & Meena, S. (2025) · Lecture notes in networks and systems · doi:10.1007/978-3-032-03740-4_19 · Cited by 3
The AI Trust Paradox: Designing Interpretable and Reliable ChatbotsPreprint
Jaidka, K., & Cai, M. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5632850 · Cited by 1
The Graph Language: How Knowledge Graphs Speak to Large Language Models
Pirrò, G. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-09527-5_33
The Value of a Sesotho Bot as Tutor for an Introductory Python Programming Class
Smit, I., & Butgereit, L. L. (2025) · Springer eBooks · doi:10.1007/978-981-95-0687-3_13
Structured Relevance Assessment for Robust Retrieval-Augmented Language Models
Garg, A. V., Raj, A., & Anitha, D. (2025) · Lecture notes in networks and systems · doi:10.1007/978-3-032-06691-6_8
Nurturing Knowledge: A Virtue Epistemology Approach to Explainable AI
Candiotto, L., Growiec, J., Kellmeyer, P., Rabinowitch, I., & Livermore, M. A. (2025) · Oxford University Press eBooks · doi:10.1093/9780198945215.003.0176
Arthur, A. W., Moore, A., Carmona-Rosado, L., Kassam, P., Logel, M., Nieves, C. I., Tope, P., Litaker, D. G., & Borrell, L. N. (2025) · Preventive Medicine · doi:10.1016/j.ypmed.2025.108436
DSMR-SQL: Enhancing Text-to-SQL with Dual-Strategy SQL Generation and Multi-role SQL Selection
Huang, Y., Guo, J., Zeng, J., Gao, C., Han, P., & Liu, C. (2025) · Lecture notes in computer science · doi:10.1007/978-981-95-2725-0_26
Zheng, L., Xu, Q., Ma, J., & Yong, W. (2025) · International Conference on Generative Artificial Intelligence for Business · doi:10.1145/3766918.3766939
“Despite their significant capabilities, LLMs frequently encounter substantial limitations in novelty assessments due to reliance on implicit reasoning processes and susceptibility to hallucinations—generating factually incorrect or exaggerated novelty claims [5], [6].”
APIDocBooster: An Extract-Then-Abstract Framework for Augmenting API Documentation
Yang, C., Liu, J., Xu, B., Treude, C., Lyu, Y., He, J., Li, M., & Lo, D. (2025) · IEEE International Conference on Software Maintenance (ICSM) · doi:10.1109/icsme64153.2025.00014 · Cited by 15
“This phenomenon, known as hallucination [23].”
Ding, Y., & Inoue, T. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-10156-3_12 · Cited by 1
Enhancing Hallucination Detection in Large Language Models by Internal-External Feature Fusion
Wang, Q., Liu, F., Li, L., & Wang, J. (2025) · International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC) · doi:10.1109/cyberc66434.2025.00039
“The rapidly developing Large Language Models (LLMs) in recent years such as GPT-4 [1], LLama [2], GLM [3], have demonstrated remarkable performance in a variety of natural language processing (NLP) tasks.”
Knowledge Discovery from Scholarly Resources Using Large Language Model and Knowledge Graph
Li, J., Xian, G., Huang, Y., & Sun, T. (2025) · IEEE International Conference on Big Data Analytics (ICBDA) · doi:10.1109/icbda65366.2025.11211295
“Despite their success in many applications, LLMs have substantial limitations when confronted complex knowledge reasoning tasks, failing to provide accurate answers to questions requiring specialized knowledge, raising concerns about responsibility, explainability and transparency [7].”
Yu, T., Wang, H., Li, S., Luo, Y., Xian, L., He, H. X., Gong, H., Zhu, T., & Ye, G. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5704125
Self-citationHealthRAG: Knowledge-Driven Medical QA System
Tuan, N. M., Cuong, T. B., Hung, L. V., Nhat, T. H., & Hung, P. D. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-10486-1_50
Automatic Evaluation of Hallucinations in Large Language Models with Semantic Inference
Castro, T., & Fernández-Isabel, A. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-10486-1_35
GRAIL: Guided Retrieval via Layer-Wise Discrepancy from Extrapolated Final Distributions
Liu, K., Zhang, H., Liu, Z., & Zhou, S. (2025) · Lecture notes in computer science · doi:10.1007/978-981-95-4384-7_8
A System Architecture for AI-Driven Market Entry Strategy Generation Using Large Language Models
Scherz, D., Ortega, J. A., Wieland, M., et al. (2025) · Procedia Computer Science · doi:10.1016/j.procs.2025.10.033
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.




