Newest papers

Major Revision, Minor Accountability: LLMs as Reviewers of Published Articles in Top-Tier Journals

Bastien Le Guellec, Su Hwan Kim, Raphaël Bentegeac, Aghilès Hamroun · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.31414 · cited by 0 works (OpenAlex) · open-access PDF

Hindi

विशेष रूप से वैज्ञानिक आउटपुट के विकास के साथ, पीएचआर समीक्षा प्रणाली एक टूटने के बिंदु पर पहुंच रही है: समीक्षकों को सुरक्षित करने के लिए तेजी से मुश्किल हो रही है, और टूटने का समय इसके अनुरूप लंबा हो रहा है। एलएलएम-सक्षम या एलएलएम-जीनरेटेड पीएचआर समीक्षा का एक उपाय के रूप में प्रस्तावित किया गया है, लेकिन इस तरह के मॉडल पहले से ही शीर्ष स्तर के पत्रिकाओं में प्रकाशित मैनुअल का इलाज करेंगे, वर्तमान प्रणाली को वैज्ञानिक वैधता के एक सोने के मानक के करीब है, यह ज्ञात नहीं है। यहां हमने 200 लेखों का एक कॉर्पस इकट्ठा किया है, प्रत्येक एक विशिष्ट CC BY 4.0 लाइसेंस के तहत जारी किया गया है और

Machine translation (Meta M2M100); the English original follows.

Abstract

With the growth of scientific output, the peer review system is approaching a breaking point: reviewers are increasingly hard to secure, and turnaround times lengthen accordingly. LLM-assisted or LLM-generated peer review has been proposed as a remedy, but how such models would treat manuscripts already published in top-tier journals, the closest thing the current system has to a gold standard of scientific validity, is not known. Here we assembled a corpus of 200 articles, each released under a CC BY 4.0 license and published in 2026, from 20 general-science, life-science, and clinical-medicine journals including Nature, Science, and Cell, and asked four current LLMs (two proprietary, two open-weight) to produce independent peer-review reports of each one under a blinded protocol. Their decisions were overwhelmingly harsh and starkly inconsistent with each other: 605 of 800 reviews (75.6%) recommended major revision or rejection, but the rate at which any single model did so ranged from 17.5% to 100% (p < 0.001), disagreement on manuscripts that had, by construction, already survived real editorial review at the journals in question. When the same manuscripts were doctored with one deliberately planted, field-specific methodological or statistical flaw, detection rates varied from 17.0% to 60.5% across models (p < 0.001). Recommending revision for papers that had already passed editorial review (low specificity) while missing flaws deliberately placed in front of them (low sensitivity), these models produce output that should be read for what it is: an uneven, framing- and model-dependent signal.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Minor (academic), Publishing, License, Impact factor, History, Original research

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange