AI tool detects LLM-generated text in research papers and peer reviews
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
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?
A recent Nature news report on Pangram's deployment at AACR notes that a sizeable minority of peer-review reports and a smaller but non-trivial fraction of manuscripts contain detectable LLM text, almost always without disclosure.
“In small scale testing, a recent Nature news report on Pangram's deployment at the American Association for Cancer Research notes that a sizeable minority of peer-review reports and a smaller but non-trivial fraction of manuscripts contain detectable LLM text, almost always without disclosure 9 .”
From Fine-Grained Detection of AI-Generated Writing in the Biomedical Literature (2026) · cites this paper as [9]
Cites
Technical Report on the Pangram AI-Generated Text Classifier
Emi, B., & Spero, M. (2024) · arXiv · doi:10.48550/arxiv.2402.14873 · Cited by 2 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
The Epistemic Downside of Using LLM-Based Generative AI in Academic Writing
Tang, B. L. (2025) · Publications · doi:10.3390/publications13040063 · Cited by 16 · see citing works
Fine-Grained Detection of AI-Generated Writing in the Biomedical Literature
She, R. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.01.01.697311 · Cited by 4 · see citing works
Large language models for peer review in biotechnology
Li, W., Zhao, R., Sun, P.-T., Lee, J., Lu, O., Yuan, S., Sureshkumar, P., Chen, Y., Xiao, Z., & Tang, Y. J. (2026) · New Biotechnology · doi:10.1016/j.nbt.2026.03.007 · Cited by 4 · see citing works
The Use and Perceptions of AI Chatbots in Medical Research: An International Cross-Sectional Survey
Alturaiki, H. M., Khamees, M. M. A., Alradhi, H. A., Alharbi, O. T., Kefi, F. Z., Alali, G. A., Bello, F., Shahzad, T., & Ramadan, A. H. A. (2026) · Cureus · doi:10.7759/cureus.100908 · Cited by 2 · see citing works
Foo, C. Y. S., Potter, K. W., Wright, A. C., et al. (2025) · Journal of Mental Health · doi:10.1080/09638237.2025.2607990 · Cited by 2 · see citing works
“While current AI detection technologies are imperfect and should not be the sole determinants of acceptance or rejection, they can serve as a useful screening mechanism to flag submissions warranting additional editorial attention (Naddaf, 2025; Zhuang et al., 2025).”
Tang, S. (2025) · 2025 8th Asia Conference on Cognitive Engineering and Intelligent Interaction (CEII) · doi:10.1109/ceii67787.2025.00040
“This development in neural network models, trained and democratized on high-quality natural language processing systems, enables researchers to use deep learning algorithms for a wide variety of computational problems, such as automated literature analysis and intelligent data mining [1].”
All Accept, No Reject: Evaluating LLMs as “Peer” Reviewers
Verma, N., & Landrum, A. R. (2026) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3772318.3791300
- Authors
- Naddaf, M.
- Corresponding author
- Naddaf, Miryam
- Venue
- Nature
- Published
- 2025-09-11
- ISSN
- 0028-0836 (ISSN-L)
- 1476-4687
- OpenAlex
- W4414120875
- Cited by
- 7
- References
- 1
- Citations by year
- 2026: 4 · 2025: 3
- Type
- other
- Language
- en
- Open access
- closed
- Licence
- Locations
- Nature (publishedVersion)
- PubMed (publishedVersion)
- Topics
- Social Sciences › Safety Research › Academic integrity and plagiarism
- Physical Sciences › Computer Science › Artificial Intelligence › Authorship Attribution and Profiling
- Social Sciences › Decision Sciences › Information Systems and Management › Academic Publishing and Open Access
- Keywords
- large language models, LLM-generated text detection, research integrity, peer review, author disclosure, editorial policy, AI in scientific publishing
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.
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
