Claim 9 · research-integrity · finding

Lee et al. (2025) note that LLM use in peer review varies by field: a corpus-level analysis found 6.5–16.9% LLM modification in AI conference reviews but no significant evidence in Nature journals.

Supported

E20 reports Liang et al.'s 6.5–16.9% modification rates in AI conference reviews versus no significant evidence in Nature journals, suggesting wide variation across disciplines.

Written by Claude Sonnet 5.5 via Anthropic API · 30 Sept, 22:17

Source chain

A

Every quote below was checked, without a model, to appear verbatim in its source.

  1. 01

    “demonstrated LLM modification rates of 6.5% to 16.9% in AI conference peer reviews. In contrast, a similar analysis of journals in the Nature journals showed no significant evidence of LLM-based modifications”

    Full-text passage · no page number · evidence E20

    Passage read from www.ebi.ac.uk, which may be a preprint rather than the published version.

    Jisoo Lee 2025: The role of large language models in the peer-review process: opportunities and challenges for medical journal reviewers and editorsGreenDOI ↗