一项对冰岛大学设计的旨在回答有关欧盟问题的AI服务Evrópuvefur的评估研究发现，虽然开放式网络搜索提供了更广泛的覆盖范围，但其引用的来源也更容易出现不信任或不相关的情况。相反，精选知识库提供了更高的可信度，但范围有限。研究强调，来源的可信度是AI信息服务中一个关键但常常被忽视的方面，尤其是在面向公众的应用中。 AI
影响 强调了在面向公众的人工智能信息服务中进行可靠来源验证的必要性，以维持用户信任。
排序理由 该集群包含一篇详细介绍AI信息服务研究结果的学术论文。
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
arXiv:2607.05217v1 Announce Type: cross Abstract: Public institutions increasingly use large language models (LLMs) to answer citizens' questions, often pairing a curated knowledge base with live web search, yet whether the sources behind these answers can be trusted has received…
Public institutions increasingly use large language models (LLMs) to answer citizens' questions, often pairing a curated knowledge base with live web search, yet whether the sources behind these answers can be trusted has received little empirical scrutiny. We report a pre-launch…
Public institutions increasingly use large language models (LLMs) to answer citizens' questions, often pairing a curated knowledge base with live web search, yet whether the sources behind these answers can be trusted has received little empirical scrutiny. We report a pre-launch…