Survey of Hallucination in Natural Language Generation
Open accessThe DOI matches its Crossref record; 355 of the 360 references it cites were found in a registry or an index, and 4 are not checked yet; 224 of the 256 sources its claims cite were checked for support.
Abstract
Natural Language Generation (NLG) has improved exponentially in recent years thanks to the development of sequence-to-sequence deep learning technologies such as Transformer-based language models. This advancement has led to more fluent and coherent NLG, leading to improved development in downstream tasks such as abstractive summarization, dialogue generation, and data-to-text generation. However, it is also apparent that deep learning based generation is prone to hallucinate unintended text, which degrades the system performance and fails to meet user expectations in many real-world scenarios. To address this issue, many studies have been presented in measuring and mitigating hallucinated texts, but these have never been reviewed in a comprehensive manner before. In this survey, we thus provide a broad overview of the research progress and challenges in the hallucination problem in NLG. The survey is organized into two parts: (1) a general overview of metrics, mitigation methods, and future directions, and (2) an overview of task-specific research progress on hallucinations in the following downstream tasks, namely abstractive summarization, dialogue generation, generative question answering, data-to-text generation, and machine translation. This survey serves to facilitate collaborative efforts among researchers in tackling the challenge of hallucinated texts in NLG.
Verified 2026-11-19 · 4 references not checked yet
Overview
CoChat Summary
TL;DR
This survey provides a broad overview of hallucination in natural language generation, covering definitions, contributors, metrics, mitigation methods, and task-specific progress across summarization, dialogue, GQA, data-to-text, machine translation, vision-language generation, and LLMs. It distinguishes intrinsic from extrinsic hallucination and argues that hallucination must be treated differently across tasks and sources.
Key findings
- The survey categorizes hallucination into intrinsic hallucination, where generated output contradicts the source, and extrinsic hallucination, where output cannot be verified from the source; extrinsic hallucination is not always erroneous because it may come from factually correct external information.
- It reports prevalence examples from cited work: 25% of summaries generated by state-of-the-art abstractive summarization systems have hallucinated content; 62% of first sentences in WIKIBIO have additional information not stated in the corresponding infobox; and 74% of samples in the QMSum dataset consist of inconsistent facts.
- It reviews hallucination metrics as statistical, model-based (IE-based, QA-based, NLI-based, faithfulness classification, LM-based), and human evaluation, and mitigation methods as data-related and modeling/inference methods including architecture changes, planning/sketching, RL, multi-task learning, controllable generation, and post-processing.
- For LLMs, it notes that RLHF reduced the hallucination rate from 41% to 21% in closed-domain tasks for InstructGPT, and discusses reference-dependent and reference-free hallucination metrics plus mitigation via data refinement, RAG, safety fine-tuning, model editing, decoding methods, chain-of-thought variants, post-processing, and ensembling.
Methods
- Literature survey design: the paper reviews existing research on hallucination in NLG rather than conducting new experiments or primary data collection.
- Coverage: general NLG hallucination topics (definitions, contributors, metrics, mitigation, future directions) plus task-specific sections on abstractive summarization, dialogue generation, generative question answering, data-to-text generation, neural machine translation, vision-language generation, and large language models.
- Analysis: synthesizes and compares definitions, metrics, mitigation approaches, and future directions across tasks, with examples in tables and figures; the LLM section was updated in Jan 2024.
Limitations
- The paper states that GQA has no standard definition or categorization of hallucination and no automatic metric specifically for hallucination in GQA; it also notes that open-domain dialogue lacks a standard hallucination metric and that VL hallucination research is still in its early stage.
- The paper states that extrinsic hallucination has been somewhat overlooked and is more challenging to reduce, and that most existing mitigation methods do not distinguish between intrinsic and extrinsic hallucination.
- Evident from the text: as a survey, it does not provide new empirical validation of metrics or mitigation methods, and its LLM coverage is explicitly updated only through Jan 2024, so later work is not included.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures
Verification
Claims the paper makes
124 Valid · 26 Not valid · 19 Not checked · 135 Own results and interpretations
At least one cited source doesn’t support this claim or points to a different paper, and none confirmed it — see the verdicts below.
In the paper
“Furthermore, perturbation, where the insertions of most common tokens are placed at the beginning of the sentence, seems to be the most successful in hallucination mitigation.”
What the sources say
- Not checked — no full text of the sourcecites [133] [133] Katherine Lee, Orhan Firat, Ashish Agarwal, Clara Fannjiang, and David Sussillo. 2019. Hallucinations in Neural Machine Translation. ICLR (2019). -
- Disputedcites [215] [215] Vikas Raunak, Arul Menezes, and Marcin Junczys-Dowmunt. 2021. The Curious Case of Hallucinations in Neural Machine Translation. (4 2021), 1172-1183. https://arxiv.org/abs/2104.06683v1 - sourceExperimental — accuracy not yet measured
In the passage describing the perturbation procedure (page unavailable), the paper says that inserting common tokens at the beginning is used to generate hallucinations, not mitigate them. Its prevention section instead discusses approaches such as robust learning and corpus-level filtering. (LLM, conf=0.92)
Checked by a reasoning model — may be wrong.
hedgedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
At least one cited source doesn’t support this claim or points to a different paper, and none confirmed it — see the verdicts below.
In the paper
“While there are works on the visual extension of faithfulness-oriented RLHF [303], these essentially involve labeling more data, which is expensive and difficult to scale.”
What the sources say
- Disputedcites [303] [303] Tianyu Yu, Yuan Yao, Haoye Zhang, Taiwen He, Yifeng Han, Ganqu Cui, Jinyi Hu, Zhiyuan Liu, Hai-Tao Zheng, Maosong Sun, and Tat-Seng Chua. 2023. RLHF-V: Towards Trustworthy MLLMs via Behavior Alignment from Fine-grained Correctional Human Feedback. CoRR abs/2312.00849 (2023). https://doi.org/10.48550/ARXIV.2312.00849 arXiv:2312.00849 - sourceExperimental — accuracy not yet measured
RLHF-V does use human annotation for visual-model alignment, but it makes the contrary point about needing more labeled data. The Fine-grained Correctional Human Preference Collection section says its corrections improve data efficiency, and Section 4.3 reports comparable hallucination rates using only 200 preference examples versus an order of magnitude more ranking-labeled examples. Our copy does not establish a separate claim about annotation cost or scalability. (LLM, conf=0.88)
Checked by a reasoning model — may be wrong.
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Disputed
Every source the paper cites for this claim was checked and supports it.
In the paper
“Nevertheless, as illustrated by Falke et al. [66], off-the-shelf NLI models tend to transfer poorly to the abstractive summarization task.”
What the sources say
- Supportedcites [66] [66] Tobias Falke, Leonardo FR Ribeiro, Prasetya Ajie Utama, Ido Dagan, and Iryna Gurevych. 2019. Ranking Generated Summaries by Correctness: An Interesting but Challenging Application for Natural Language Inference. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics . 2214-2220. - source
The Introduction states that models trained on NLI datasets transfer poorly to the paper’s downstream task of ranking generated summaries by correctness. The Conclusions likewise say that out-of-the-box NLI models do not perform well on that task. (LLM, conf=0.94)
- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Such wrong association results in generation with facts mixed up between two similar entities [60, 233].”
What the sources say
- Partly supportedcites [60] [60] Nouha Dziri, Andrea Madotto, Osmar Zaiane, and Avishek Joey Bose. 2021. Neural Path Hunter: Reducing Hallucination in Dialogue Systems via Path Grounding. EMNLP (2021). - sourceExperimental — accuracy not yet measured
The GPT2-KG passage describes an incorrect association between entities, and G Error Analysis gives an example of a generated response assigning a book to the wrong author. These support the claim that facts can be mixed up, but the paper does not say the entities involved are similar. The supplied passages do not provide page numbers. (LLM, conf=0.83)
- Supportedcites [233] [233] Kurt Shuster, Spencer Poff, Moya Chen, Douwe Kiela, and Jason Weston. 2021. Retrieval Augmentation Reduces Hallucination in Conversation. EMNLP (2021). - source
The Introduction states that language models often mix up facts between two similar entities when generating factually incorrect statements. This directly supports the cited description of the resulting error. (LLM, conf=0.92)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“As a result, the target reference may contain information that cannot be supported by the source [198, 268].”
What the sources say
- Supportedcites [198] [198] Ankur Parikh, Xuezhi Wang, Sebastian Gehrmann, Manaal Faruqui, Bhuwan Dhingra, Diyi Yang, and Dipanjan Das. 2020. ToTTo: A Controlled Table-To-Text Generation Dataset. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) . 1173-1186. - source
The Dataset section says that reference targets selected from web sentences may contain information that cannot be inferred from the source. The Data Annotation Process section also describes deleting phrases unsupported by table cells. (LLM, conf=0.93)
- Supportedcites [268] [268] Hongmin Wang. 2019. Revisiting Challenges in Data-to-Text Generation with Fact Grounding. In Proceedings of the 12th International Conference on Natural Language Generation . 311-322. - source
The Introduction reports that about 40% of human-written game-summary content cannot be directly mapped to the input boxscore records and draws on information beyond those inputs. Section 2.1 likewise describes reference content not licensed by the boxscore table. This supports the claim that a target reference may contain information the source does not support. (LLM, conf=0.93)
hedgedstates a possible cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
117 more claims · 74 with their cited sources checked
Create a free account to see all 304 claims and to verify the 2 references still open.
- Borderline: one of the tests for a substantial claim is unclear. 1
- It may state two separate claims as one. 90
- Probably not a substantial claim. 142
- The quote only partly states this claim. 1
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Hallucination can lead to serious consequences such as incorrect or misleading code, security risks, and ethical issues [186].”
From A survey on large language model based autonomous agents (2024) · cites this paper as [186]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“citation fabrication constitutes a form of extrinsic hallucination in the absence of retrieval augmentation, since the model generates references not grounded in any provided source material [4].”
From How LLMs Cite and Why It Matters: A Cross-Model Audit of Reference Fabrication in AI-Assisted Academic Writing and Methods to Detect Phantom Citations (2026) · cites this paper as [4]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Recent taxonomic work distinguishes at least seven cases [15,18].”
From Hallucination as epistemic pollution: epistemic injustice and credibility in futures-oriented science education (2026) · cites this paper as [18]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“LLMs struggle to generate accurate and up-to-date responses on current topics, due to outdated knowledge, lack of domain-specific expertise, or hallucination13–17.”
From Leveraging long context in retrieval augmented language models for medical question answering (2025) · cites this paper as [16]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Fine-tuning can enrich LLMs with domain knowledge, enhance their specific ability, improve the fairness and reliability of the outputs, and prevent certain damage caused by hallucination [26].”
From FederatedScope-LLM: A Comprehensive Package for Fine-tuning Large Language Models in Federated Learning (2024) · cites this paper as [26]
Experimental — accuracy not yet measured
41 more claims about this paper
Create a free account to see every claim other papers make about this one.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Extrinsic Object Hallucination
“CHAIR (Caption Hallucination Assessment with Image Relevance) metric, which calculates what proportion of object words generated are actually in the image according to the ground truth captions.”
§ Extrinsic Object Hallucination
“CHAIR 𝑖 measures per-instance object hallucination, i.e. what fraction of object instances in each generated caption are hallucinated.”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ 7.1 Hallucination Definition in Abstractive Summarization
“For instance, in Table 1, given the input article shown in the caption, an example of intrinsic hallucination is ' The Ebola vaccine was rejected by the FDA in 2019, '”
§ 2.1 Categorization
“in the abstractive summarization task from Table 1, the generated summary ' The first Ebola vaccine was approved in 2021 ' contradicts the source content”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Survey of Hallucination in Natural Language Generation
“The survey is organized into two parts: (1) a general overview of metrics, mitigation methods, and future directions; (2) an overview of task-specific research progress on hallucinations in the following downstream tasks, namely abstractive summarization, dialogue generation, generative question answering, data-to-text generation, machine translation, and visual-language generation; and (3) hallucinations in large language models (LLMs)”
§ 1 INTRODUCTION
“The third part discusses this phenomenon in LLMs in Section 13.”
Citations
Cites
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325 · cited 3×
“Recently, the rapid development of NLG has captured the imagination of many thanks to the advances in deep learning technologies, especially Transformer [189]-based models like BERT [29], BART [100], GPT-2 [149], and GPT-3 [16].”
BackgroundAn Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., et al. (2020) · arXiv · doi:10.48550/arxiv.2010.11929 · Cited by 47,472
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
Sequence to Sequence Learning with Neural Networks
Sutskever, I., Vinyals, O., & Le, Q. V. (2014) · arXiv · doi:10.48550/arxiv.1409.3215 · Cited by 17,766
Advances in Neural Information Processing Systems 29
Teymur, O., Zygalakis, K., & Calderhead, B. (2016) · Cited by 13,887
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., Zhou, Y., Li, W., & Liu, P. J. (2019) · arXiv · doi:10.48550/arxiv.1910.10683 · Cited by 10,929 · cited 1×
“This work investigates the performances of GPT-3 [13], GPT-Neo/J [141], GPT-2 [105] and a T5-based model [106].”
BackgroundAdvances in Neural Information Processing Systems 14: Proceedings of the 2001 Conference
(2002) · The MIT Press eBooks · doi:10.7551/mitpress/1120.001.0001 · Cited by 8,890
Lewis, M., Liu, Y., Goyal, N., Ghazvininejad, M., Mohamed, A., Levy, O., Stoyanov, V., & Zettlemoyer, L. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.703 · Cited by 8,821 · cited 1×
“Recently, the development of NLG has conquered the imagination of many thanks to the advances in deep learning technologies, especially the Transformer [159]-based models like BERT [24], BART [83], GPT-2 [127], and GPT3 [13].”
BackgroundLewis, M., Liu, Y., Goyal, N., Ghazvininejad, M., Mohamed, A., Levy, O., Stoyanov, V., & Zettlemoyer, L. (2019) · arXiv · doi:10.48550/arxiv.1910.13461 · Cited by 8,816 · cited 1×
“Recently, the development of NLG has conquered the imagination of many thanks to the advances in deep learning technologies, especially the Transformer [159]-based models like BERT [24], BART [83], GPT-2 [127], and GPT3 [13].”
Background
Research works printed but not linked: 16
- [4] Joshua Albrecht and Rebecca Hwa. 2007. A Re-examination of Machine Learning Approaches for Sentence-Level MT Evaluation. In Proceedings of the 45th Annual Meeting of the Association of Computational Linguistics . Association for Computational Linguistics, Prague, Czech Republic, 880-887. https://aclanthology.org/P07-1111 -
- [16] Andrzej Białecki, Robert Muir, Grant Ingersoll, and Lucid Imagination. 2012. Apache lucene 4. In SIGIR 2012 workshop on open source information retrieval . 17. -
- [53] Yilun Du, Shuang Li, Antonio Torralba, Joshua B Tenenbaum, and Igor Mordatch. 2023. Improving Factuality and Reasoning in Language Models through Multiagent Debate. In Forty-first International Conference on Machine Learning . -
- [121] Philipp Koehn and Rebecca Knowles. 2017. Six Challenges for Neural Machine Translation. In First Workshop on Neural Machine Translation . Association for Computational Linguistics, 28-39. -
- [145] Junnan Li, Dongxu Li, Caiming Xiong, and Steven Hoi. 2022. BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation. In ICML . -
- [149] Tian Li, Ahmad Beirami, Maziar Sanjabi, and Virginia Smith. 2020. Tilted Empirical Risk Minimization. In International Conference on Learning Representations . -
- [164] Jiasen Lu, Jianwei Yang, Dhruv Batra, and Devi Parikh. 2018. Neural Baby Talk. In CVPR . -
- [165] Yukun Ma, Khanh Linh Nguyen, Frank Z Xing, and Erik Cambria. 2020. A Survey on Empathetic Dialogue Systems. Information Fusion 64 (2020), 50-70. -
- [175] Marianna J. Martindale, Marine Carpuat, Kevin Duh, and Paul McNamee. 2019. Identifying Fluently Inadequate Output in Neural and Statistical Machine Translation. In MTSummit . -
- [179] Kathleen McKeown. 1992. Text Generation . Cambridge University Press. -
- [209] Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, and Ilya Sutskever. 2019. Language Models are Unsupervised Multitask Learners. OpenAI Blog 1, 8 (2019), 9. -
- [238] Hui Su, Xiaoyu Shen, Sanqiang Zhao, Zhou Xiao, Pengwei Hu, Cheng Niu, and Jie Zhou. 2020. Diversifying Dialogue Generation with Non-Conversational Text. In 58th Annual Meeting of the Association for Computational Linguistics . ACL, 7087-7097. -
- [256] Zhaopeng Tu, Yang Liu, Lifeng Shang, Xiaohua Liu, and Hang Li. 2017. Neural Machine Translation with Reconstruction. In Thirty-First AAAI Conference on Artificial Intelligence . -
- [261] Neeraj Varshney, Wenlin Yao, Hongming Zhang, Jianshu Chen, and Dong Yu. 2023. A Stitch in Time Saves Nine: Detecting and Mitigating Hallucinations of LLMs by Validating Low-Confidence Generation. CoRR abs/2307.03987 (2023). https://doi.org/10.48550/arXiv.2307.03987 arXiv:2307.03987 - · arXiv · doi:10.48550/arxiv.2307.03987 · Paper page
- [285] Zeqiu Wu, Michel Galley, Chris Brockett, Yizhe Zhang, Xiang Gao, Chris Quirk, Rik Koncel-Kedziorski, Jianfeng Gao, Hannaneh Hajishirzi, Mari Ostendorf, et al. 2021. A Controllable Model of Grounded Response Generation. In Proceedings of the AAAI Conference on Artificial Intelligence , Vol. 35. 14085-14093. -
- [291] Xinnuo Xu, Ondřej Dušek, Verena Rieser, and Ioannis Konstas. 2021. AGGGEN: Ordering and Aggregating while Generating. roceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL2021) (2021). -
Cited by
Items 3,001–3,100 of 4,844
How scenarios can impact small models for QA task in education?
Tarifa, A., & Chabir, C. (2025) · Procedia Computer Science · doi:10.1016/j.procs.2025.09.525 · Cited by 2
Towards Trustworthy Conversational Agents: A New Method for Hallucination Detection in NLG
Ghani, S., Håkansson, A., & Pasichnyi, O. (2025) · Procedia Computer Science · doi:10.1016/j.procs.2025.09.651
Evaluating Generative AI and Human Performance in Question-Answer Validation Tasks
Opaiţ, E.-E., Duca, A., & Olariu, M.-E. (2025) · Procedia Computer Science · doi:10.1016/j.procs.2025.09.522
Can AI Match Professional Analysts? Evidence from a Multi-Agent SystemPreprint
Francis, A., & Zhang, C. X. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5584131
Exploring Causal Effect of Social Bias on Faithfulness Hallucinations in Large Language Models
Zhang, Z., Zhang, J., Hu, X., Zhang, H., & Wan, X. (2025) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3746252.3761298
From Menus to the Interactive Food-Ordering Systems
Kim, M.-J., Park, S.-J., Ha, J., Seo, J. Y., Aliyeva, D., & Kim, K.-M. (2025) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3746252.3761001
C-FAITH: A Chinese Fine-Grained Benchmark for Automated Hallucination Evaluation
Zhang, X., Liu, Z., Wang, J., Zhang, H., Xu, F., Zhang, J., & Wan, X. (2025) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3746252.3761604 · Cited by 3
Guo, J., Zeng, X., Wu, J., & Li, Z. (2025) · Communications in computer and information science · doi:10.1007/978-981-95-4088-4_15
Chen, Y., Zhang, R., Liu, Y., Zhu, T., Liu, M., & Lv, W. (2025) · Communications in computer and information science · doi:10.1007/978-981-95-4088-4_8
Neuss, C. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5722752
MPCR-SQL: Enhancing SQL Generation by Multi-path SQL Generation and Collective SQL Refinement
Hou, M., Huang, Y., & Wang, S. (2025) · Communications in computer and information science · doi:10.1007/978-981-95-4091-4_15
LLiMe: enhancing text classifier explanations with large language models
Angiulli, F., Luca, F. D., Fassetti, F., & Nisticò, S. (2025) · Machine Learning · doi:10.1007/s10994-025-06920-z · Cited by 6
“Hallucinated behaviours can be split into the following two groups (Ji et al., 2023).”
KE-LPG: Toward Semantic Refinement for Lesson Plan Generation
Guo, H., Zhu, J., Pan, C., Wang, Q., Zhou, C., & Meng, C. (2025) · Arabian Journal for Science and Engineering · doi:10.1007/s13369-025-10808-4 · Cited by 3
A Multi-source Temporal Graph Approach for Reliable Market Forecasting with LLM Synergy
Shi, Y. (2025) · Lecture notes in computer science · doi:10.1007/978-981-95-3061-8_19
ChatGPT use in dissertations by postgraduates
Singh, T., & Malik, R. (2025) · BDJ · doi:10.1038/s41415-025-9334-7
Zhang, B., Lu, J., Liang, Y., Xu, B., & Lin, H. (2025) · Lecture notes in computer science · doi:10.1007/978-981-95-3349-7_39 · Cited by 1
Benchmarking Retrieval-Augmented Generation: Challenging Domain-specific Datasets Generation Method
Liu, Y., Hu, L., Dong, X., & Wang, Z. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11228599 · Cited by 2
“However, these models still face numerous challenges, including hallucination of facts [7] and lack of domain-specific expertise [20].”
From One Bit to Meaningful Self: Recursive Identity Formation for Decentralized Digital Trust
Stankovski, V. (2025) · International Conference on Blockchain Computing and Applications (BCCA) · doi:10.1109/bcca66705.2025.11229562
“Consequently, they may achieve syntactic plausibility, but fall short of autonomous, reality-based knowledge discovery [5], [23].”
Cognitive Support in Aviation Operations Utilizing Multimodal Large Language Models
Huang, R., Feng, S., Li, F., Chen, C., & Ong, Y.-S. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11229131 · Cited by 1
“Never-theless, MLLMs lack the specialized operational knowledge necessary for aviation and may produce hallucinations, e.g., outputs that seem accurate but lead to incorrect responses [10], which can be hazardous during flight.”
Dynamic Cognitive Bias: Hallucination and Forgetting in the Cognitive Dynamics of LLMs
Ni, Z., Tang, S., Guo, X., Sun, L., & Lv, P. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11229003
“This issue has been widely observed in various language modeling tasks, such as summarization, translation, and question answering [14], [31].”
Working Towards Toxic datasets for LLM Safeguarding
Guo, L., Wang, Z., Wang, Z., & He, T. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11227669
“As LLMs become more capable, risks from misalignment—such as hallucination [33], sycophancy [34], and deception [35]—have gained attention.”
Detecting and Correcting Hallucinations in LLMs via Substantive Uncertainty and Iterative Validation
Chen, Z., Cheng, Y., & Gao, Y. (2025) · Lecture notes in computer science · doi:10.1007/978-981-95-3352-7_14 · Cited by 1
Barbato, F., Sorichetti, F., Ielo, E., et al. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-12092-2_23
Shi, J., Zhang, X., Qu, X., Xie, J., & Wang, J. (2025) · Frontiers of Information Technology & Electronic Engineering · doi:10.1631/fitee.2500282 · Cited by 1
Ddog: optimizing multi-hop inference via dual-driven retrieval and reasoning path
Chen, Y., Gu, B., Gao, L., Fu, K., Qu, Y., & Cui, L. (2025) · Machine Learning · doi:10.1007/s10994-025-06917-8
“LLMs are prone to generating hallucinations, producing content that appears plausible but is not factually accurate (Ji et al., 2023; Xu et al., 2025).”
Detecting Token-Level Hallucinations Using Variance Signals: A Reference-Free Approach
Kumar, K. (2025) · Conference of Open Innovations Association (FRUCT) · doi:10.23919/fruct67853.2025.11239326
Probing the reasoning abilities of LLMs in blocks world
Zhao, K. Z., & Macbeth, J. (2025) · Cognitive Systems Research · doi:10.1016/j.cogsys.2025.101421
Explaining Hallucination of Artificial Intelligence in Data-Driven Decision-Making
Kabir, Z. S. (2025) · IGI Global eBooks · doi:10.4018/979-8-3373-5702-7.ch004 · Cited by 1
Lahmy, M., & Yozevitch, R. (2025) · arXiv · doi:10.48550/arxiv.2512.10787 · Cited by 1
“• Judge Variance: Although we mitigate bias using a blind protocol (judges see only retrieved passages) and fixed rubrics, LLM-based evaluation remains subject to stochasticity [12].”
Impact of Artificial Intelligence on the Strategic Decision Making of Multinational Companies
Macleod, B., & Pugalia, S. (2025) · IGI Global eBooks · doi:10.4018/979-8-3373-5702-7.ch007
Smarter, not Bigger: Fine-Tuned RAG-Enhanced LLMs for Automotive HIL TestingPreprint
Feng, C., Liu, Z., Gupta, S., Cui, G., von der Assen, J., & Stiller, B. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5781609 · Cited by 5
Elbaghazaoui, B. E., Moudden, T. E., & Benabbes, K. (2025) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-5616-7.ch009
Von der Polygraphie zur KI-basierten Wahrheitsfindung
Preuß, T. (2025) · Springer eBooks · doi:10.1007/978-3-658-48425-5_9
Navigating the future of teaching, learning, and creativity with generative AI
Acar, S., & Ellis, A. (2025) · Elsevier eBooks · doi:10.1016/b978-0-443-34073-4.00003-4
Bi-directional Semantic Alignment for Large Vision Language Model Training via Echo-Former
Wang, Y., Liu, Q., Li, Z., Wang, S., Guo, S., & Wei, Z. (2025) · Lecture notes in computer science · doi:10.1007/978-981-95-3346-6_6
Parenti, S. I., Maglioni, A., Evangelisti, E., Gracco, A. L. T., Badiali, G., Bonetti, G. A., & Bartolucci, M. L. (2025) · Applied Sciences · doi:10.3390/app152312441 · Cited by 1
Xie, Z., Chan, T.-T., & Yu, P. L. H. (2025) · Computers and Education Artificial Intelligence · doi:10.1016/j.caeai.2025.100510 · Cited by 1
Lin, H., Gao, W. G., Ma, J., Deng, Y., Luo, Z., Wang, B., Yang, R., & Chua, T. (2025) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2025.3637552 · Cited by 4
“Nonetheless, there exists a considerable risk that rationales derived directly from LMMs may be influenced by pre-existing biases, potentially resulting in inaccurate classification and adversely affecting the detection efficacy [45], [46].”
Psychological Assistance Model Based on Fine-Tuning and Knowledge Base Technology
Wu, J.-H., Liu, X.-M., Peng, J.-F., Peng, H., & Liao, H. (2025) · 2025 International Conference on Computational Intelligence and Robotics (CIR) · doi:10.1109/cir65373.2025.11257373
“This validates the RAG mechanism’s effectiveness: by retrieving DSM-5-TR and crisis intervention knowledge in real time, the model avoids "hallucinations" common in LLMs [29] .”
Lin, J., Yang, Q., Li, Y., Dong, K., & Liu, H. (2025) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros60139.2025.11246507 · Cited by 1
“To address the hallucination issues [21] of large language models (LLMs) [1] in domain-specific applications, retrieval-augmented generation (RAG) enhances LLM outputs through semantic retrieval from external knowledge bases [17].”
Sim, W. S., Lim, K. Y., & Lim, T. M. (2025) · IEEE International Conference on Computation, Big-Data and Engineering (ICCBE) · doi:10.1109/iccbe65177.2025.11255978
Zhang, N., Ye, R., & Han, J. (2025) · IEEE International Symposium on Parallel and Distributed Processing with Applications (ISPA) · doi:10.1109/ispa67752.2025.00159
“However, these models remain prone to hallucination errors [4], manifesting as factual inaccuracies [5], [6] and an inherent inability to guarantee output veracity [7], [8].”
Is our chatbot telling lies? Assessing correctness of an LLM-based Dutch support chatbot
Lassche, H., Overeem, M., & Rastogi, A. (2025) · Journal of Systems and Software · doi:10.1016/j.jss.2025.112699 · Cited by 1
Enhancing static vulnerability alert validation using large language models
Chen, S.-S., Hsu, Y.-S., Lin, T.-C., et al. (2026) · Journal of Systems and Software · doi:10.1016/j.jss.2026.113040
Augmenting Large Language Models with Dense Retrieval Techniques for Enhanced QA Performance
Shetty, S., Shetty, S., & Menezes, J. (2025) · IEEE International Conference on Distributed Computing, VLSI, Electrical Circuits and Robotics (DISCOVER) · doi:10.1109/discover66922.2025.11258922
“Even with advancements, successfully incorporating retrieved information into responses is still difficult because hallucinations — coherent but ungrounded outputs — continue to erode user confidence [1], [2].”
Enhancing SAP Predictive Maintenance and Industrial Analytics with LLM-Driven Automation
Chandrakumaran, A. (2025) · IEEE Madhya Pradesh Section Conference (MPCON) · doi:10.1109/mpcon66082.2025.11256577 · Cited by 6
Automated Public Announcement System Using Template-based Natural Language Generation
Yora, S., & Ramadhan, M. M. (2025) · International Conference on Information Management and Technology (ICIMTech) · doi:10.1109/icimtech67074.2025.11265561
“Using a deep learning approach tends to generate hallucinating text that raises unwanted text, which degrades system performance and fails to meet user expectations in many real-world scenarios, so Template-based NLG is more reliable [22].”
Generation of Score Description Sentences For School Report Cards Using NLG
Yora, S., & Ramadhan, M. M. (2025) · International Conference on Information Technology and Computing (ICITCOM) · doi:10.1109/icitcom66635.2025.11265387
“Besides using deep learning approaches that can lead to hallucinating text and out-of-context [18], [19], [20], the application of those methods is different compared to the problem of this research. b) Template-based NLG: All of template-based NLG is usually used to generate sentences with…”
Towards enhancing data science agents with semantics
Hoseini, S., Ibbels, M., Knoll, M., & Quix, C. (2025) · Computer Science and Information Systems · doi:10.2298/csis250320078h
Yoshikawa, M., Miuzuno, T., Ohto, Y., Fujimoto, H., & Kusuhara, H. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2025.12.01.691726
Integrating Generative AI into Translator Training: Challenges and Opportunities
Kornacki, M. (2025) · AUC PHILOLOGICA · doi:10.14712/24646830.2025.12 · Cited by 4
NeuReg: Neuro-Symbolic QA Generation from Regulatory Compliance
Arshad, M. U., Corsar, D., & Nkisi-Orji, I. (2025) · International Conference on Knowledge Capture · doi:10.1145/3731443.3771375 · Cited by 2
Fleissner, K., Jazdi, N., & Zhou, D. (2025) · IFAC-PapersOnLine · doi:10.1016/j.ifacol.2025.11.873
Gamification Empowered by LLM-Based Agents: A Systematic Literature Review
Boubakri, M., & Nafil, K. (2025) · International Conference on Intelligent Systems: Theories and Applications · doi:10.1109/sita67914.2025.11273696
“LLM agents can hallucinate, generating fluent but outright false responses [45], and often appear to lack successful transparency of their decision-making rationale.”
Information Literacy in the Age of Large Language Models
Bewersdorf, B., & Wittingslow, R. (2025) · Springer eBooks · doi:10.1007/978-3-031-99804-1_6 · Cited by 1
Approach to Eye Tracking Scanpath Analysis with Multimodal Large Language Model
Li, X., Yin, K., & Gu, Y. (2025) · Modelling—International Open Access Journal of Modelling in Engineering Science · doi:10.3390/modelling6040164
“Consequently, MLLMs struggle with fine-grained visual parsing, which leads to missed or misinterpreted details in complex gaze sequences [37].”
Sowing trust in the machine: ethical pathways for large language models in farmer-centered extension
Kabir, K. H. (2025) · AI & Society · doi:10.1007/s00146-025-02788-8 · Cited by 4
“They predict likely continuations of prompts, which can cause hallucinations—confident but incorrect answers (Zhang et al. 2026; Ji et al. 2023).”
Chatzimina, M. E., & Tsiknakis, M. (2025) · IEEE International Conference on Bioinformatics and Bioengineering (BIBE) · doi:10.1109/bibe66822.2025.00061
“However, when applied to highly specialized domains such as hematologic malignancies, their outputs often suffer from factual inaccuracies, known as hallucinations [3].”
Understanding the Impact of Dataset Characteristics on RAG-based Multi-hop QA PerformancePreprint
Aksoy, N., Güven, Z. A., & Ünalır, M. O. (2025) · Research Square · doi:10.21203/rs.3.rs-6968562/v2
Levi‐Belz, Y., Nobile, B., Levkovich, I., Courtet, P., & Elyoseph, Z. (2026) · European Psychiatry · doi:10.1016/j.chbr.2026.101189
Han, H. W., Kim, J. H., Chio, J., Shim, S. R., Lee, J., & Lee, M. (2025) · doi:10.2196/preprints.87065
A red teaming framework for large language models: a case study on faithfulness evaluation
Alotaibi, A., Mughus, R., & Ahmed, M. (2026) · Software Quality Journal · doi:10.1007/s11219-026-09779-y · Cited by 4
Wibowo, J. N., Shidik, G. F., Mareta, A. P. D., & Kusuma, E. J. (2025) · International Seminar on Application for Technology of Information and Communication (iSemantic) · doi:10.1109/isemantic67418.2025.11291991
“While this transition enables more abstractive and natural responses, it also introduces a critical failure mode: hallucination, the generation of plausible-sounding content that is factually incorrect or ungrounded in the provided source text [1].”
Luo, X., Huo, Y., Xiao, Y., & Yu, X. (2025) · International Conference on Artificial Intelligence, Automation and High Performance Computing (AIAHPC) · doi:10.1109/aiahpc66801.2025.11290242
“However, the use of low-intelligence language recognisers can lead to ’hallu-cination’ errors in these models [6], as they struggle with factual inaccuracies [13], [14] and lack the ability to ensure the accuracy of the content generated [15], [28].”
Ghani, S., Håkansson, A., Pasichnyi, O., & Shahrokni, H. (2025) · IOP Conference Series Earth and Environmental Science · doi:10.1088/1755-1315/1546/1/012027 · Cited by 1
“Building on prior work in conversational agents [4] and RAG [6], SPARA addresses challenges like hallucinations [7], stability , and reliability [9], enhancing precision by integrating expert knowledge from energy advisors.”
Research on Enterprise Security Operation Based on Large Model Technology
Li, L., Wang, T., An, Q., et al. (2025) · International Conference on Computer Information Science and Artificial Intelligence (CISAI) · doi:10.1145/3773365.3773614
Consistency retrieval enhancement method for large language models for power system fault analysis
Li, W., Su, B., Wang, N., Tian, Y., Huang, X., & He, T. Y. (2025) · International Symposium on Artificial Intelligence and Computational Social Sciences (AICSS) · doi:10.1145/3776759.3776840
“Accurate understanding of user needs directly determines the quality of retrieved content; (2) Large language models are prone to hallucinations [7], and the risk increases as dialogue length grows.”
Wu, K., Yuhuan, C. H. E. N., Xiao, W., & Jiawei, H. U. (2025) · Research Square · doi:10.21203/rs.3.rs-8037100/v1
Farhan, Y. H., Tareq, M., Shannaq, B., Almaqbali, S., & Ali, O. (2025) · International Conference on Cyber Resilience (ICCR) · doi:10.1109/iccr67387.2025.11291967 · Cited by 1
Overcoming Challenges in AI-Powered NLP Models: Enhancing the Capabilities of ChatGPT and DeepSeek
Shannaq, B., Farhan, Y. H., Tareq, M., Ali, O., & Almaqbali, S. (2025) · International Conference on Cyber Resilience (ICCR) · doi:10.1109/iccr67387.2025.11291926
Adain, A. A., Alasadi, A. Y., Hameed, M., Mehd, R., Hamza, A. A., Hameed, D. W., & Mustafa, A. (2025) · International Conference on Cyber Resilience (ICCR) · doi:10.1109/iccr67387.2025.11292048
Balancing Engagement and Polarization: Multi-Objective Alignment of News Content Using LLMsPreprint
Cheng, M., Ofek, E., & Yoganarasimhan, H. (2025) · RePEc: Research Papers in Economics · doi:10.48550/arxiv.2504.13444 · Cited by 7
LLM-Rank: An Unsupervised Keyword Extraction Method Using Local Large Language Models
Jia, X., Roller, C., & Wang, C. (2025) · IEEE International Conference on Information Reuse and Integration (IRI) · doi:10.1109/iri66576.2025.00021 · Cited by 1
“However, challenges persist: hallucination generates irrelevant keywords not grounded in the text, and prompt sensitivity leads to inconsistent outputs based on input phrasing [21].”
A Technical Framework For Formally Verifying AI Models Supporting 6G Network Functions
Mahmud, M. M. H., Baby, R., Moazzeni, S., Parra-Ullauri, J. M., Zhou, X., Wu, Y., Katsaros, K., Mavromatis, I., Zeb, S., Hussain, R., et al. (2025) · IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS) · doi:10.1109/infocomwkshps65812.2025.11152991
“Using AI models to support 6G network functions [1] is challenging in this context, as they are inherently black-box, complex, and have vulnerabilities like adversarial noise and hallucinations [2]– [4].”
O que são as “alucinações” da Inteligência Artificial
Choi, A., & Mei, K. (2025) · The Conversation · doi:10.64628/ade.g7nwjxcaf
Enhancing User Alignment in Large Language Models via Counterfactual Context Divergence DecodingPreprint
Wang, Y., Wang, J., & Luo, Z. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5644037
From Essay to AI-Augmented Writing: Reframing Pedagogy and Feedback Through the AAWP ModelPreprint
Askari, M., & Rahim, A. S. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5771957
Hou, W., Zhuang, Y., Liu, Y., & Liu, X. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.130473 · Cited by 2
Künstliche Intelligenz in der Netzwerkforensik
Spiekermann, D. (2025) · Springer eBooks · doi:10.1007/978-3-658-48425-5_11
Security Risks in Large Language Models and General Mitigation Strategies
Zhang, R., & Kanyane, M. H. (2025) · IFIP advances in information and communication technology · doi:10.1007/978-3-032-13075-4_1
Chatzimina, M. E., Tachos, N. S., Fotiadis, D. I., & Tsiknakis, M. (2025) · IEEE-EMBS International Conference on Biomedical and Health Informatics (BHI ...) · doi:10.1109/bhi67747.2025.11269561
“Although recent studies highlighted the performance of LLMs, they often generate factually incorrect responses which is called hallucinations [6].”
Leveraging Large Language Models for Log Anomaly Detection: Case Studies and Comparative Analysis
Xiong, Y. (2025) · ITM Web of Conferences · doi:10.1051/itmconf/20258001039 · Cited by 2
Latent linguistics: the conception of language in large language models
Poibeau, T. (2025) · Ubiquity Press eBooks · doi:10.5334/bde.b
Ellerm, A., Adams, B., & GAHEGAN, M. N. (2025) · Open Conference Proceedings · doi:10.52825/ocp.v8i.3174 · Cited by 1
Fitrah, A. M., & Husin, N. (2025) · International Conference on Cyber and IT Service Management (CITSM) · doi:10.1109/citsm67730.2025.11291316
“We did not evaluate targeted adversarial attacks or diffusion-era composites at scale; both are planned for future work [33].”
Friend, J., Chow, C., Sideman, A. B., Pinheiro‐Chagas, P., Brooks-Smith-Lowe, S., Dulaney, S., Sawyer, R. J., Hess, M., Lum, H. D., Olney, N. T., et al. (2025) · doi:10.2196/preprints.88543
Ashraf, H. N., & Uversky, V. N. (2025) · Cell Biochemistry and Biophysics · doi:10.1007/s12013-025-01978-x
““Hal-lucination” is a new kind of error characteristic for the generative artificial intelligence (AI) models producing outputs that are highly detailed and informationally rich [ 163 , 164].”
AI-DeFi Systemic Risk: A Complex Adaptive System Framework
Zhou, Y., & Papageorgiou, O. (2025) · Conference on Blockchain Research & Applications for Innovative Networks and Services (BRAINS) · doi:10.1109/brains67003.2025.11302949 · Cited by 1
“Their nonlinear architectures (e.g., non-additive activation functions, hierarchical feedback, compositional depth) yield outputs unexpected from initial rules or rewards [89]– [91].”
Learning From the Machine’s Mistakes: Embedding AI Error Detection Into Primary Care Safety Culture
Jerjes, W., & Majeed, A. (2025) · Journal of Patient Safety · doi:10.1097/pts.0000000000001453
Queryome: orchestrating retrieval, reasoning, and synthesis across biomedical literature
Punuru, P., Ibtehaz, N., Giri, S. J., Srirangam, H., Tugolukova, E. A., & Kihara, D. (2026) · Journal of Biomedical Semantics · doi:10.1186/s13326-026-00366-5 · Cited by 3
HUMAN–AI COLLABORATION IN REVIVING FOLK TRADITIONS
Faisal, M., Ojha, V., Mithun Kumar, S., Gandhi, A., Manisha, Aggarwal, M., & Deshpande, L. (2025) · ShodhKosh Journal of Visual and Performing Arts · doi:10.29121/shodhkosh.v6.i3s.2025.6790
Using Large Language Models to Create Educational Materials for the “Databases” Discipline
Asenchik, A. D. (2025) · Digital Transformation · doi:10.35596/1729-7648-2025-31-4-5-14 · Cited by 1
Al-Muhammed, M. J. (2025) · Knowledge-Based Systems · doi:10.1016/j.knosys.2025.115199 · Cited by 1
Ghosh, S., & Raman, R. (2025) · Research Square · doi:10.21203/rs.3.rs-8401830/v1
Sharma, M., Ganie, K. A., Panwar, P., & Arti (2025) · International Journal of Latest Technology in Engineering Management & Applied Science · doi:10.51583/ijltemas.2025.1412000008
Ngoc, T., Nam, B. K. H., Viet, H. V., Linh, D. V., & Khoe, N. V. (2025) · Journal of Military Science and Technology · doi:10.54939/1859-1043.j.mst.208.2025.136-142
Research on AI Agent-Driven Enterprise Multi-Scenario Integrated System
Tang, L., Chen, R., & Li, X. (2025) · International Conference on Artificial Intelligence and Intelligent Information Processing (AIIIP) · doi:10.1145/3778534.3778672 · Cited by 2
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.




