arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2604.08602cs.DLcs.AIcs.LG
收藏

TiAb Review Plugin:一种基于浏览器的AI辅助标题和摘要筛选工具

TiAb Review Plugin: A Browser-Based Tool for AI-Assisted Study Selection in Systematic Reviews

Yuki Kataoka, Masahiro Banno, Michihito Kyo, Shuri Nakao, Tomoo Sato, Shunsuke Taito, Tomohiro Takayama, Takahiro Tsuge, Yasushi Tsujimoto, Ryuhei So, Toshi A. Furukawa

AI总结:

本文提出一种基于浏览器的无代码、无服务器AI辅助标题和摘要筛选工具,结合大语言模型和机器学习主动学习,实现了高效的系统综述筛选。

摘要:

背景:基于服务器的筛选工具会带来订阅费用,而开源替代方案需要编程技能。目标:我们开发了一个浏览器扩展,提供无代码、无服务器的人工智能(AI)辅助标题和摘要筛选,并检验其功能。方法:TiAb Review Plugin 是一个开源的 Chrome 浏览器扩展(可在 https://chromewebstore.google.com/detail/tiab-review-plugin/alejlnlfflogpnabpbplmnojgoeeabij 上获取)。它使用 Google Sheets 作为共享数据库,不需要专用服务器,能够实现多审稿人协作。用户自行提供 Gemini API 密钥,存储在本地并加密。该工具提供三种筛选模式:手动审核、大型语言模型(LLM)批量筛选和机器学习(ML)主动学习。对于 ML 评估,我们重新实现了默认的 ASReview 主动学习算法(TF-IDF 与朴素贝叶斯)在 TypeScript 中,以实现浏览器内执行,并通过 10 折交叉验证在六个数据集上验证其等效性与原始 Python 实现。对于 LLM 评估,我们在基准数据集上比较了 16 种参数配置,两个模型家族,然后在五个公开数据集(1,038 到 5,628 条记录,0.5 到 2.0% 的患病率)上验证最佳配置(Gemini 3.0 Flash,低思考预算,TopP=0.95)的敏感性导向提示。结果:TypeScript 分类器在所有六个数据集上都产生了与原始 ASReview 100% 一致的前 100 名排名。对于 LLM 筛选,召回率为 94 到 100%,精确率为 2 到 15%,工作节省率(WSS@95)在 95% 召回率范围内从 48.7 到 87.3%。结论:我们开发了一个功能齐全的浏览器扩展,将 LLM 筛选和 ML 主动学习整合到无代码、无服务器的环境中,已准备好在系统综述筛选中实际使用。

英文摘要:

Server-based screening tools impose subscription costs, while open-source alternatives require coding skills, and full-text screening has remained outside the scope of no-code open-source tools. We developed TiAb Review Plugin, an open-source Chrome browser extension that provides no-code, serverless artificial intelligence (AI)-assisted study selection covering both title and abstract (T&A) screening and full-text screening. It uses Google Sheets as a shared database and Google Drive as a PDF store, and users supply their own large language model (LLM) API key. For T&A screening, it offers manual review, LLM batch screening, and machine learning (ML) active learning. For full-text screening, it retrieves open-access PDFs from PubMed Central, Europe PMC, Unpaywall, OpenAlex, and publisher pages, supports blinded dual review with structured exclusion reasons and adjudication, optionally obtains an LLM judgment with page-anchored evidence, and computes PRISMA 2020 flow counts. We re-implemented the default ASReview algorithm (TF-IDF with Naive Bayes) in TypeScript and compared it with the Python original using 10-fold cross-validation on six datasets. For LLM T&A screening, we compared 16 parameter configurations on a benchmark dataset, validated the best (Gemini 3.0 Flash, low thinking budget, TopP 0.95) on five public datasets (1,038 to 5,628 records; 0.5% to 2.0% prevalence), and benchmarked nine further models from four developers. The TypeScript classifier produced top-100 rankings identical to ASReview on all six datasets. LLM T&A screening achieved recall of 94% to 100% with precision of 2% to 15%, and work saved over sampling at 95% recall (WSS@95) of 46.3% to 89.3%. No additional model exceeded the 96.1% recall of the reference configuration; the most recent models traded recall for precision. The classification accuracy of the full-text stage has not yet been evaluated.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑