# Kurate preprint demonstrates evidence-quality scoring across 4,347 papers

_Published Wednesday, October 7, 2026 at 12:10 AM EDT · AI, Science · Latest · Tier 2 — Notable_

Researchers report in an arXiv preprint that Kurate, a system using large language models, assessed study quality across 4,347 papers, including 3,913 reporting randomized trials. It scores eight dimensions of study design and reporting, drawing on papers and related documents such as trial registrations and protocols, and links judgments to supporting passages.

Against expert annotations of 60 held-out clinical-trial documents, extracted information matched expert labels in 221/242 protocol scorepoints and 294/370 results-publication scorepoints. Across the larger corpus, issues appeared most often in statistical power, selective reporting and analysis prespecification. Average quality differed between clinical areas.

## Sources

- [cs.CL updates on arXiv.org](https://arxiv.org/abs/2610.07306)

---
Canonical: https://techandbusiness.org/newswire/APbZpaQiHwvsbNUj2yDxhc
Published: 2026-10-07T04:10:19.639Z
Story chronology: 2026-10-07T04:00:00.000Z
Retrieved: 2026-10-07T06:14:11.530Z
Publisher: Tech & Business (techandbusiness.org)