arXiv cs.AI· Russell Ro, Reza Abbasi-Asl·· 2 天前AI 评分33
RECITE：用文本证据自动修订临床试验入组标准
RECITE: Revising Eligibility Criteria Incorporating Textual Evidence
AI 导读
RECITE 框架可从协议修订文档中自动发现、提取并落实基于证据的入组标准修订，构建了覆盖 5,735 项试验、11,913 条样本的数据集。在 3,116 条样本上评测 15 种 LLM 配置，最佳开源模型 Gemma 2 9B 与最佳闭源模型 GPT-4o 的归一化序数得分分别为 0.742 和 0.745。智能体文献发现系统另在五篇出版物中识别出 44 组论文与试验匹配。
正文 · 原文
Abstract:Overly restrictive eligibility criteria contribute to widespread enrollment failure in clinical trials, yet published recommendations for how specific criteria should be broadened remain disconnected from trial design at scale. We introduce RECITE (Revising Eligibility Criteria Incorporating Textual Evidence), a framework for automated discovery, extraction, and implementation of evidence-based eligibility criteria revisions. RECITE uses protocol amendment documents from this http URL to construct a dataset of 11,913 curated instances across 5,735 trials, each pairing original criteria with documentary evidence and target revisions. We benchmark fifteen LLM configurations under truncated-to-context-window and retrieval-augmented generation conditions, on a 3,116-instance sample under an LLM-as-judge protocol validated against ten domain experts (judge vs. human quadratic-weighted Cohen's $\kappa_{qw} = 0.694 \pm 0.096$). The best open-weight model, Gemma 2 9B, achieves $\kappa_{qw}$-aligned ordinal score $3.0 \pm 0.9$ on a 0--4 scale (normalized 0.742) and the best closed-weight model, GPT-4o, reaches $3.0 \pm 0.8$ (normalized 0.745). Using an agentic literature discovery system, we then identified 44 paper--trial matches across five publications providing trial-design and EC-modification evidence. Our findings suggest that existing clinical knowledge can be operationalized into actionable trial design recommendations.
| Comments: | 18 pages, 4 figures, 4 tables. Extended version of a paper in KDD 2026 (32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining). Code and data: this https URL |
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.41368 [cs.AI] |
| (or arXiv:2610.41368v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.41368 arXiv-issued DOI via DataCite (pending registration) |
|
| Related DOI: | https://doi.org/10.1145/3770855.3819027
DOI(s) linking to related resources |
Submission history
From: Russell Ro [view email]
[v1]
Fri, 30 Oct 2026 07:34:50 UTC (314 KB)
来源：arXiv cs.AI · arxiv.org