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 4,501–4,600 of 4,844
TVDM-LLM: Schema-Aware Natural Language Querying for Transportation Surveillance Video Databases
Chen, X., & Zhang, C. (2026) · IEEE International Conference on Information Reuse and Integration (IRI) · doi:10.1109/iri69576.2026.00111
“The formal schema grounds and constrains LLM outputs [25], while the LLM handles natural language translation.”
A comparative evaluation of LLM-generated and clinician-written psychotherapy session summaries
Keren, L., Klein, A., Bar‐Shachar, Y., Liakata, M., & Atzil–Slonim, D. (2026) · Psychotherapy Research · doi:10.1080/10503307.2026.2729093
Explainable AI Question-Answering Framework: Integrating RAG with Knowledge Graph Reasoning
Suman, R., & Barisua, B. S. (2026) · IEEE International Conference on Information Reuse and Integration (IRI) · doi:10.1109/iri69576.2026.00095
“Faithfulness was assessed using established hallucination detection metrics [11], [15], specifically the ratio of answer claims directly supported by retrieved passages or KG triples.”
van Genderen, P. J. J., & van Sprang, E. N. M. (2026) · Journal of Travel Medicine · doi:10.1093/jtm/taag082
Biancofiore, G. M., Palma, D. D., Pomo, C., Boratto, L., Noia, T. D., & Narducci, F. (2026) · Computer Science Review · doi:10.1016/j.cosrev.2026.101058
Huang, J.-L., Zhu, J., & Ni, X. (2026) · Journal of Medical Internet Research · doi:10.2196/97007
“For LLMs specifically, medical use raises concerns about fluent but unsupported outputs [21], and hallucination is a recognized limitation of natural language generation systems [22].”
Plausible nonsense and deliberative reasoning: Benchmarking LLMs against human judgment
Veri, F., & Umbelino, G. K. (2026) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2600126123
Kumar, V. (2026) · IEEE International Conference on Information Reuse and Integration (IRI) · doi:10.1109/iri69576.2026.00093
“Where natural language hallucinations embed incorrect propositions within grammatically valid text [1], structured output hallucinations violate the constraint systems that define what constitutes a valid instance of a domain.”
Patault, E., Sinsin, T., Bruis-Gervasone, L., Quesnot, P., Girard, S., Pesquier, B., & MARDUEL, B. (2026) · AI for Engineering · doi:10.3390/aieng1020011
Munnee, M. M., Malumi, O., Chinedu, S., Okah, W., Ogboye, A., Abiakam, C., Duys, T., Lagos, K. J., Etuk, I., & Mobisson, N. (2026) · Journal of Medical Internet Research · doi:10.2196/86578
Zhao, Y., Wang, Y., Pan, Y., Chen, M., Zhu, Z., Lu, Z., Chen, S., & Chen, K. (2026) · Applied Sciences · doi:10.3390/app16188940
Yin, C., & Zhang, Y. (2026) · Research Square · doi:10.21203/rs.3.rs-10758357/v1
Li, M., Bu, F., Zhang, Y., & Sun, F. (2026) · Sustainability · doi:10.3390/su18189277
Операциональная концепция истины: верификация предположений в взаимодействии с реальностьюPreprint
Петрийчук, Н. (2026) · doi:10.24108/preprints-3116338
Skill-Based Architecture for Reusable, AI-Native Enterprise Modules: An Empirical Study
Jathar, N. (2026) · IEEE International Conference on Information Reuse and Integration (IRI) · doi:10.1109/iri69576.2026.00012
“LLMs are known to produce fluent but incorrect output [17], and an enterprise system cannot assume its runtime is always accurate.”
Aparicio, E. M., Gutierrez, F. X. M., Smolyak, D., Mane, H., Yue, X., Srikanth, N. P., He, X., Nguyen, T. T. X., Boyd-Graber, J. L., Dipankar, P., et al. (2026) · JMIR Formative Research · doi:10.2196/101898
Rad, F. N. E., Pirehbabi, S., Gavagsaz‐Ghoachani, R., & Phattanasak, M. (2026) · National Conference on Technical Education and International Conference on Technical Education (NCTechEd & ICTechEd) · doi:10.1109/icteched69022.2026.11682857
Tedeschi, L. O., Greer, N., Kaniyamattam, K., Mudireddy, D. R., Strong, R., Surabhi, P. S. M., Tao, J., & Wang, A. (2026) · Frontiers in Nutrition · doi:10.3389/fnut.2026.1815038
“These strategies include retrieval augmentation, a method that introduces an external retriever into the system to reduce conversational hallucinations (46).”
Stephen, S., & Vignesh, R. (2026) · Computers & Electrical Engineering · doi:10.1016/j.compeleceng.2026.111514
Combining Formal Reasoning and LLMs for Scenario-Based Educational STEM Exercises
Vollmer, L., Loubet, R., Neider, D., Hasse, H., & Leitte, H. (2026) · Human-Centric Intelligent Systems · doi:10.1007/s44230-026-00174-1
Hanson-DeFusco, J. (2026) · Social Sciences · doi:10.3390/socsci15090616
Moganadas, S. R., Marín-González, F., Nun, S. H., & Gan, C. L. (2026) · Frontiers in Education · doi:10.3389/feduc.2026.1856839 · Cited by 1
Agha, Y., & Khan, M. S. A. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-8167-1.ch004
Noymanee, J., & Nilsook, P. (2026) · National Conference on Technical Education and International Conference on Technical Education (NCTechEd & ICTechEd) · doi:10.1109/icteched69022.2026.11682822
“…literature frames hallucination as both a technical issue (generation under uncertainty) and a socio-technical issue (misplaced user trust, inadequate verification workflows, and unclear accountability), emphasizing that high-stakes adoption requires safeguards beyond prompt-level fixes [1] – [3].”
Machine-readable immune ontogeny maps for AOP-anchoring: A critical review
Pamanji, R. (2026) · Computational Toxicology · doi:10.1016/j.comtox.2026.100449
Dai, Y., Ma, Q., Zhang, W., & Lei, H. (2026) · International Conference on Computer Information Science and Application Technology (CISAT) · doi:10.1109/cisat71030.2026.11682464
“This reduces frequency bias and hallucination, which can map partial risk signals to dominant vulnerability types [8], [9].”
Clinical warrant in medical LLM evaluation
Sariyar, M. (2026) · Frontiers in Digital Health · doi:10.3389/fdgth.2026.1918724
Zhang, W., Feng, Z., Liu, Y., Zhong, R., Chen, S., Li, Z., 李冰燕, Liu, D., Gao, L., & Hu, K. (2026) · Frontiers in Oncology · doi:10.3389/fonc.2026.1915669
da Cunha, M. V., SILVEIRA, M. R., Santana, B. S., de Freitas, L. A., & Corrêa, U. B. (2026) · Journal on Interactive Systems · doi:10.5753/jis.2026.8133
Development of an LLM-Based Multi-Agent System for Sewing Sequence Planning
Park, H., & Kim, S. (2026) · Clothing and Textiles Research Journal · doi:10.1177/0887302x261487373
Complexity-Aware Adaptive Routing for RAG: A Data-Efficient Framework for Domain-Specific Analytics
Yan, S., Hua, L., & Wang, K. (2026) · International Conference on Cloud Computing and Big Data Analytics (ICCCBDA) · doi:10.1109/icccbda69573.2026.11676501
“We employ a unified prompt template with strict grounding constraints (“Answer solely based on context”) and a deterministic decoding strategy (temperature=0) to minimize hallucination [11].”
Więckowski, J., Paradowski, B., Kizielewicz, B., & Bouraima, M. B. (2026) · Applied Soft Computing · doi:10.1016/j.asoc.2026.116417 · Cited by 1
OceanSkill: An Agentic Skill Framework for Evidence-Grounded Ocean Forecast Report Generation
Yu, X., Bi, Z., Zhu, Y., Wang, Z., Liu, Z., & Lou, J. (2026) · International Conference on Large Models and Intelligent Agent Technologies (LMIAT) · doi:10.1109/lmiat70705.2026.11682146
“Advances in large language models (LLMs) provide new opportunities for automatic scientific report generation [12], [13] and LLMs may produce reports that are fluent but incomplete, unsupported, or physically inconsistent [14].”
Zhang, C. (2026) · Open Computer Science · doi:10.1515/comp-2025-0073
“In academic writing scenarios, hallucination may appear as fabricated references, unsupported factual claims, or misleading methodological statements [12], [13].”
Meridian topology-aware graph-language alignment for acupuncture prescription generation
Zhang, W., Cui, J., Zhang, M., Zhao, M., Mu, Y., & Li, P. (2026) · Neurocomputing · doi:10.1016/j.neucom.2026.135086
A Formal Approach to LLM Hallucination Detection and Mitigation
Wang, Z., Liu, Y., & Song, W. W. (2026) · International Conference on Cloud Computing and Big Data Analytics (ICCCBDA) · doi:10.1109/icccbda69573.2026.11676510
“However, when generating content like project reports, business analyses, and technical documents, LLM tools i often produce “hallucinations” — seemingly plausible outputs that, in reality, do not align with factual information[2].”
TACO: Trust Assessment of Large Language Models in Coding Assistance Tasks
Weng, S., Feng, Y., Li, J., Yin, Y., Zhang, Z., Liu, L., & Liu, J. (2026) · ACM/IEEE International Conference on Software Engineering: Companion Proceedings (ICSE-Companion) · doi:10.1145/3744916.3787835
Cherniaeva, P. D., Kukartsev, V. A., Prudkiy, A. S., et al. (2026) · doi:10.1109/icsmartgrid70882.2026.11680779
“Extensive taxonomies of hallucinations in generative models prove that these networks often generate internally coherent but factually ungrounded outputs due to probabilistic token sampling [18-19].”
Jiang, H., Liu, J., & Gong, H. (2026) · Asia Pacific Journal of Marketing and Logistics · doi:10.1108/apjml-04-2026-0860
Zhao, Z., Zhang, S., Jin, X., et al. (2026) · Applied Energy · doi:10.1016/j.apenergy.2026.128819
Auditable LLM-assisted decision support for sustainable infrastructure planning
Elkliny, A., Abdulwasea, T. K., Ibrahim, A. S., Deng, X., & Gaber, M. A. (2026) · Proceedings of the Institution of Civil Engineers - Engineering Sustainability · doi:10.1680/jensu.26.00094 · Cited by 1
A Systematic Review of Modern Machine Learning in Query Optimization
Limnaios, E., & Kondylakis, H. (2026) · ACM Computing Surveys · doi:10.1145/3847509
Carruthers, Y. A., & Heim, J. (2026) · AI & Society · doi:10.1007/s00146-026-03365-3
Şimşek, A. R., & Tantug, A. C. (2026) · Bilgisayar bilimleri ve mühendisliği dergisi. · doi:10.54525/bbmd.1850314
Pal, A. (2026) · Discover Artificial Intelligence · doi:10.1007/s44163-026-02240-w
“Broader surveys of hallucination in natural language generation and LLMs show that fluent generation can be unfaithful to source material, inconsistent with world knowledge, or unsupported by available evidence [2, 3].”
Agentic AI workflows in anatomic pathology: Opportunities and challenges
Ngo, A., Tian, R., Mohammad, A., Weon, J. L., Arvisais‐Anhalt, S., Minor, J., & Youens, K. (2026) · Journal of Pathology Informatics · doi:10.1016/j.jpi.2026.100718
AI Literacies for Adults in a GenAI Intensive Workforce: A Thematic Literature Review (2022–2026)
Creely, E. (2026) · Journal of Adolescent & Adult Literacy · doi:10.1002/jaal.70075
Zeng, Y., Lou, P., Ming, T., Gao, R., & Hu, J. (2026) · Machines · doi:10.3390/machines14091044
“Generated content may consequently be unsupported or inconsistent with domain facts, representing a concrete industrial risk of large-language-model hallucination [2].”
Alviz-Meza, A., Valencia-Arias, A., Rojas-Flores, S., & Díaz, F. (2026) · Processes · doi:10.3390/pr14182917 · Cited by 1
Klapproth, F. (2026) · Education Sciences · doi:10.3390/educsci16091507 · Cited by 2
Schwarz, M., & Hashemi, D. (2026) · Kardiologie up2date · doi:10.1055/a-2761-8099
Paulett, J. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-32732-1_18
Which Words to Teach: Word Selection in the Age of Large Language Models
Hiebert, E. H. (2026) · Encyclopedia · doi:10.3390/encyclopedia6090201
“The model can propose words, but it cannot verify them [65].”
Language-Augmented Video Action Anticipation: Design Fundamentals, Benchmarks, and Open ChallengePreprint
Mohammadi, M., Fu, Z., & Rowlands, S. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.7460177
Guo, Z., Qiu, R., Qiu, X., Yang, Y., Wei, D., & Li, Y. (2026) · International Conference on Educational Digitalization and Computer Science in the Guangdong-Hong Kong-Macao Greater Bay Area · doi:10.1145/3828183.3828329
“They may hallucinate nonexistent policies such as a “medical waiver policy” [3].”
Ornelas, J. C. O. (2026) · Nexus Science Review · doi:10.66104/tbe64a08
“The clinically important distinction is between linguistic plausibility and epistemic grounding (Ji et al., 2023; Thirunavukarasu et al., 2023).”
Zhu, Z., Wang, Y., Yin, X., Zhu, S., & Yang, W. (2026) · International Conference on Educational Digitalization and Computer Science in the Guangdong-Hong Kong-Macao Greater Bay Area · doi:10.1145/3828183.3828358
“However, prior studies have shown that LLMs relying solely on parametric knowledge may suffer from factual inconsistency and hallucination problems [ 4 ] .”
Operationalizing Trust in AI Outputs: A Multi-signal, Task-Aware, Trust Scoring Framework
Rajaram, K., Manohar, K., Hosein, P., & Addison, L. M. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-35078-7_17
Stenzel, D. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-32729-1_30
Qi, T. (2026) · Frontiers in Psychology · doi:10.3389/fpsyg.2026.1929537
“These concerns are sharper in physical training because inaccurate or overconfident wording can encourage compensatory movement or unsafe range (Bender et al., 2021; Ji et al., 2023).”
Dejene, B., Assabie, Y., Tadele, M., Adnew, B., Tsegaye, R., Sintayehu, Y., Alemu, A., Howe, R., Kiragga, A., Girma, T., et al. (2026) · Technologies · doi:10.3390/technologies14090579
Nitchals, C. (2026) · AI and Ethics · doi:10.1007/s43681-026-01379-7 · Cited by 1
Adaptive Optics Image Analysis Using Generative AI (GPT-4) Scripting: Exploratory Study
Pas, J. A. A. H., van der Zanden, L., van Leeuwen, Y., Lechanteur, Y. T. E., Hoyng, C. B., & Theelen, T. (2026) · JMIR Formative Research · doi:10.2196/94906
“Feeding the idea of “blob detection analysis” to GPT-4 1 Feeding GPT-4 with previously generated code to stay on track 32, 39, and 43 Feeding GPT-4 with previously generated code.”
Almeshari, M. A., Asiri, O. A., Alahmari, A. D., Alghamdi, A. S., Alsaif, S. S., Alhammad, S. A., Bamogaddam, Y., Alowaisi, M., Alasmari, A., Alobaidi, N. Y., et al. (2026) · Frontiers in Medicine · doi:10.3389/fmed.2026.1935478
“Although students utilized LLMs to simplify complex concepts (82%), 75% acknowledged that these models may produce incorrect information — a well-documented phenomenon referred to as hallucination (23).”
Lu, G., & Wu, T. (2026) · International Conference on Machine Learning, Natural Language Processing and Modeling · doi:10.1145/3832928.3833023
Sani, M. S., Iranmanesh, S., & Raad, R. (2026) · Sensors · doi:10.3390/s26185862
“This issue has been examined for natural-language generation by Ji et al. [28] and specifically for large language models by Huang et al. [29].”
Bouhafs, A. (2026) · Question(s) de management · doi:10.3917/qdm.240.0065
Yuan, G. X. (2026) · International Journal of Financial Engineering · doi:10.1142/s2424786326500477
Augmenting Expertise in Organizational Foresight: The Role of GenAI in Collaborative Settings
Zumbrunn, L. (2026) · Futures & Foresight Science · doi:10.1002/ffo2.70052 · Cited by 2
He, Z., & Octavia, A. (2026) · International Conference on Digital Society and Artificial Intelligence (DSAI) · doi:10.1145/3833541.3833679
Liu, R., Pang, F., Jiang, H., Cai, Z., Zhang, M., & Li, L. (2026) · International Conference on Digital Society and Artificial Intelligence (DSAI) · doi:10.1145/3833541.3833828
Liu, X. (2026) · International Conference on Software Engineering and Computer Applications · doi:10.1145/3829641.3829712
“Drivenbyarelianceonimplicit parametricmemory,LLMsexhibitapronouncedvulnerabilityto "factualhallucinations"duringcomplexclinicalconsultations[1].”
Bi, L. (2026) · International Conference on Digital Society and Artificial Intelligence (DSAI) · doi:10.1145/3833541.3833786
“This study has achieved the following core accomplishments: First, this study proposes a dual-agent framework integrating a baseline question-answering agent and a Retrieval-Augmented Generation (RAG) agent [5, 6].”
InfluentialKyurkchiev, P., & Iliev, A. I. (2026) · Future Internet · doi:10.3390/fi18090481 · Cited by 1
Intelligent Transformation of HR Payroll Data Based on Large Language Models
Wang, Y. (2026) · International Conference on Digital Society and Artificial Intelligence (DSAI) · doi:10.1145/3833541.3833656
FatPlants 2.0: an AI ‐powered platform integrating plant lipid genes, pathways, and literature
Qin, Y., Guo, C., Kim, M., Zhang, Y., Azam, M., Xiao, T., Xu, C., Shorrosh, B. S., Durrett, T. P., Allen, D. K., et al. (2026) · The Plant Journal · doi:10.1111/tpj.71117
“…and question answering (Gao, Fang, et al., 2024; OpenAI et al., 2024); however, their application to biological knowledge exploration remains limited by hallucination, lack of traceability, and insufficient grounding in structured domain knowledge (Gao, Xiong, et al., 2024; Ji et al., 2023).”
Structure‐Prior Guided Legal Judgment Summarization: Chain‐of‐Thought and Attention Backtracking
Zhao, H., Li, Y., Wang, J., & Xi, Y. (2026) · Concurrency and Computation Practice and Experience · doi:10.1002/cpe.70947
Li, W. (2026) · International Conference on Software Engineering and Computer Applications · doi:10.1145/3829641.3829736
“The autonomous extraction of information via generative models introduces massive systemic risks into knowledge graphs, prominently including hallucinations, logical contradictions, and temporal distortions [ 29 ] .”
Zaboski, B. A., Sugino, E., & King, K. (2026) · Journal of Clinical Medicine · doi:10.3390/jcm15187154 · Cited by 1
“A clinician who endorses a particular product or configuration is endorsing something that may behave differently sometime later, and current systems can generate fluent inaccuracies on clinical topics without signaling that they have done so [49,60].”
Gorga, A. L., & Verna, L. (2026) · JLIS it · doi:10.36253/jlis.it-735
“Within this perspective lies the growing interest in the adoption of Retrieval-Augmented Generation (RAG) architectures, in which the generative capabilities of large language models (LLMs) are anchored to bounded and verifiable data sources.”
InfluentialFrom Faithful to Blurred: Quantifying News Information Degradation in Multi-LLM Chain-of-Retelling
Wang, H. (2026) · International Conference on Digital Society and Artificial Intelligence (DSAI) · doi:10.1145/3833541.3833571
Boumazzourh, M., Moujahdi, C., & Ziti, S. (2026) · International Conference on Connected Innovation and Technology (ICCITX) · doi:10.1109/iccitx70146.2026.11679936
“One problem that arises often is ‘hallucination’ where the model hallucinates some information that makes sense but is ultimately not real [3].”
Gao, J., Yu, B., Chen, Y., Gao, K., Bao, S., & Wang, J. (2026) · Computer-Aided Civil and Infrastructure Engineering · doi:10.1016/j.cacaie.2026.100231
A Benchmark Dataset for Decision-Oriented Retrieval-Augmented Generation
Chen, J., Ling, Y., Fu, Y., Lu, Z., & Zhang, J. (2026) · International Conference on Big Data and Information Analytics (BigDIA) · doi:10.1109/bigdia71496.2026.11684649
“Large language models (LLMs) have made substantial progress in natural language processing; however, they still face challenges such as outdated knowledge and hallucinations in knowledge-intensive tasks [1][2] .”
A practical risk framework for large language model use in life science research
Sharpton, T., Davis II, E. W., & Alexiev, A. (2026) · PLOS Computational Biology · doi:10.1371/journal.pcbi.1014776
“Asked for a reference, it may generate citation-like text whether that citation exists or not, reproducing associations from training even where they do not apply [14,15].”
Liu, X., Jiao, P., Xie, X., Yang, S., & Ju, R. (2026) · International Conference on Big Data and Information Analytics (BigDIA) · doi:10.1109/bigdia71496.2026.11684585
Research on Local State Assessment Method Based on Complex-Task Time-Series Data
Wang, C., Zhai, W., Liao, Z., Liu, Y., Xie, Y., Wang, X., & Tang, J. (2026) · International Conference on Big Data and Information Analytics (BigDIA) · doi:10.1109/bigdia71496.2026.11684657
“…is lengthy and highly subjective; 3) when large language models are directly used for indicator design in specialized vertical domains, they are prone to issues such as knowledge hallucination and unsupported field fabrication, making it difficult to directly support reliable assessment [11], [12].”
Reading between the Tokens: Entropy as a Predictor of Hallucinations in Small-Scale Language Models
Shetty, Y., Prabhanjana, G., Das, D., et al. (2026) · International Conference on Trends in Quantum Computing and Emerging Business Technologies (TQCEBT) · doi:10.1109/tqcebt67648.2026.11681169
“Hallucination: Boolean indicator flagging responses that are factually incorrect but stated with confidence; correct responses and those that acknowledge uncertainty are not flagged [16].”
E-Commerce Chatbot with Graph RAG & LLM
Hrithika, N., & Raja, S. E. (2026) · International Conference on Trends in Quantum Computing and Emerging Business Technologies (TQCEBT) · doi:10.1109/tqcebt67648.2026.11681829
“The implementation of LLMs as separate systems enables them to produce false information because they lack access to verified knowledge bases [4].”
InfluentialJi, H., Luo, Y., Pang, M., Fang, Y., Li, J., & Qi, X. (2026) · doi:10.2196/preprints.112100
ChatGPT as a self-regulated learning support tool in thesis writing: a mixed-methods study
Annamalai, N., Niu, C., Wu, X., & Alkubaisy, A. A. A. (2026) · Frontiers in Psychology · doi:10.3389/fpsyg.2026.1871579
He, Y., Wang, L., Luo, W., Yu, X., Li, Y., Liu, D., Guo, M., Hu, S., & Zhao, J. (2026) · Sustainable civil infrastructures · doi:10.1007/978-3-032-32237-1_22
Khakimullina, D. F., & Davletbaeva, D. N. (2026) · Philology Theory & Practice · doi:10.30853/phil20260435
Cho, M., Song, J., Jin, J., Baek, G., Kim, H., Noh, B., & Tak, S. (2026) · The Journal of The Korea Institute of Intelligent Transport Systems · doi:10.12815/kits.2026.25.4.45
Rojas, H., Larraín, A. F., Sanhueza, D., Mery, D., & Dunstan, J. (2026) · Tapuya Latin American Science Technology and Society · doi:10.1080/25729861.2026.2698211 · Cited by 3
von Maltitz, M., Fabris‐Rotelli, I., Roberts, D. J., Smit, A., Maposa, D., & Das, S. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-9089-5.ch010
Artificial intelligence in toxicology: current advances, challenges and future directions
Kuča, K., Otari, S., Singh, N., Gide, M., Somani, R., Křenek, J., Palla, D., & Nepovimová, E. (2026) · Archives of Toxicology · doi:10.1007/s00204-026-04551-4
Zhou, Y., & Yu, C. (2026) · IFIP advances in information and communication technology · doi:10.1007/978-3-032-37199-7_3
Mastery-Guided Retrieval-Augmented Generation for Grounded Tutoring
Fijiwala, J. V., & Bakarola, V. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-31233-4_35
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.




