边验证边浏览:一种基于大型语言模型的假新闻检测浏览器扩展
📄 中文摘要
假新闻在数字时代的泛滥对公众信任和民主制度构成了严重威胁,因此迫切需要有效、透明且以用户为中心的检测工具。现有的浏览器扩展往往因模型行为不透明、解释支持有限以及缺乏有意义的用户参与而显得不足。Aletheia是一款新型浏览器扩展,利用检索增强生成(RAG)和大型语言模型(LLM)来检测假新闻并提供基于证据的解释。此外,Aletheia还包括两个互动组件:讨论中心,允许用户围绕标记内容进行对话,以及“保持知情”功能,展示最新的事实核查。通过广泛的用户测试,Aletheia旨在提升用户的新闻识别能力和参与感。
📄 English Summary
Verify as You Go: An LLM-Powered Browser Extension for Fake News Detection
The rampant spread of fake news poses significant risks to public trust and democratic institutions, highlighting the urgent need for effective, transparent, and user-centered detection tools. Existing browser extensions often fall short due to opaque model behavior, limited explanatory support, and a lack of meaningful user engagement. Aletheia is a novel browser extension that leverages Retrieval-Augmented Generation (RAG) and Large Language Models (LLMs) to detect fake news and provide evidence-based explanations. Additionally, Aletheia includes two interactive components: a Discussion Hub that facilitates user dialogue around flagged content and a Stay Informed feature that surfaces recent fact-checks. Through extensive user testing, Aletheia aims to enhance users' news discernment capabilities and engagement.
Powered by Cloudflare Workers + Payload CMS + Claude 3.5
数据源: OpenAI, Google AI, DeepMind, AWS ML Blog, HuggingFace 等