大模型
arxiv.org ·
审查患者教育中生成式 AI 的机构异质性：一项针对 102 份美国移植手册的大规模研究
该 arXiv 论文（cs.CY）用结构化输出 LLM judge 对 23 家美国实体器官移植中心的 102 份患者教育手册进行 1,772,261 次两两比较，并结合 1,115 个患者问题（TransplantQA）审计机构间内容一致性。主要发现：同中心跨器官一致性比跨中心同器官高 0.024（Holm 校正 p=0.011）；信息缺口集中在代表性不足的亚群体，生殖健康尤为突出（82% 缺失、86% 被判定为高显著性）；judge 主题聚为 991 类，免疫抑制与妊娠时机优先级最高；问题与覆盖特征可回溯预测高分歧问题（AUC 0.77）。论文讨论了面向患者的生成式 AI 在移植护理中的部署含义。
AI推荐理由 arXiv 预印本，作者来自卡内基梅隆大学（Rema Padman、Ramayya Krishnan 为知名学者），方法明确：用结构化输出 LLM judge 对 23 家美国实体器官移植中心、102 份患者教育手册做 1,772,261 次两两比较，并配 1,115 个患者问题（TransplantQA）。给出四项可核查发现：同中心跨器官一致性高于跨中心同器官（+0.024，Holm 校正 p=0.011）；生殖健康存在 82% 缺失、86% 高显著性；991 个主题聚类；高分歧问题可回溯预测（AUC 0.77）。局限：仅覆盖移植领域与美国机构，结论外推需谨慎；LLM judge 本身存在偏差风险，论文未在摘要中说明人工验证比例。对医疗 AI 落地与 RAG 知识库治理有直接参考价值，但属学术审计研究，非产品发布。
72
正文
Computer Science > Computers and Society
arXiv:2607.22606 (cs)
[Submitted on 13 Jun 2026 (v1), last revised 23 Sep 2026 (this version, v2)]
Title:Auditing Institutional Heterogeneity for Generative AI in Patient Education: A Large-Scale Study of 102 US Transplant Handbooks
Authors:Yubo Li, Rema Padman, Ramayya Krishnan
View a PDF of the paper titled Auditing Institutional Heterogeneity for Generative AI in Patient Education: A Large-Scale Study of 102 US Transplant Handbooks, by Yubo Li and 2 other authors
View PDF
HTML (experimental)
Abstract:Health systems are rapidly deploying generative-AI assistants that answer patient questions from institution-authored education materials, on the premise that grounding in local content yields consistent guidance. Do the underlying documents themselves agree? We use a structured-output large-language-model judge to audit 1{,}772{,}261 pairwise comparisons across 102 patient-education handbooks from 23 US solid-organ transplant centers, paired with 1{,}115 patient-derived questions (TransplantQA). Four findings bear directly on deployment: (1) same-center cross-organ agreement exceeds cross-center same-organ agreement by $0.024$ in the primary analysis (Holm-adjusted $p=0.011$), with sensitivity to document selection; (2) information gaps concern topics relevant to underrepresented subgroups, with reproductive health a \emph{double jeopardy}: 82\% absence and 86\% judge-rated high significance among divergent/contradictory pairs; (3) judge-derived themes form 991 clusters, with immunosuppression and pregnancy timing among the highest judge-rated priorities; (4) question and observed-coverage features predict high-divergence questions retrospectively (AUC $0.77$). We discuss implications for deploying patient-facing generative AI in transplant care.
Subjects:
Computers and Society (cs.CY)
Cite as:
arXiv:2607.22606 [cs.CY]
(or
arXiv:2607.22606v2 [cs.CY] for this version)
https://doi.org/10.48550/arXiv.2607.22606
Focus to learn more
arXiv-issued DOI via DataCite
Submission history
From: Yubo Li [view email]
[v1]
Sat, 13 Jun 2026 20:57:32 UTC (740 KB)
[v2]
Wed, 23 Sep 2026 11:47:48 UTC (69 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled Auditing Institutional Heterogeneity for Generative AI in Patient Education: A Large-Scale Study of 102 US Transplant Handbooks, by Yubo Li and 2 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.CY
new
|
recent
| 2026-07
Change to browse by:
cs
References & Citations
NASA ADS
Google Scholar
Semantic Scholar
export BibTeX citation
Loading...
BibTeX formatted citation
×
loading...
Data provided by:
Bookmark
Bibliographic Tools
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media
Code, Data and Media Associated with this Article
alphaXiv Toggle
alphaXiv (What is alphaXiv?)
Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub Toggle
DagsHub (What is DagsHub?)
GotitPub Toggle
Gotit.pub (What is GotitPub?)
Huggingface Toggle
Hugging Face (What is Huggingface?)
ScienceCast Toggle
ScienceCast (What is ScienceCast?)
Demos
Demos
Replicate Toggle
Replicate (What is Replicate?)
Spaces Toggle
Hugging Face Spaces (What is Spaces?)
Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
Related Papers
Recommenders and Search Tools
Link to Influence Flower
Influence Flower (What are Influence Flowers?)
Core recommender toggle
CORE Recommender (What is CORE?)
Author
Venue
Institution
Topic
About arXivLabs
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Which authors of this paper are endorsers? |
Disable MathJax (What is MathJax?)