<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>LLM脑腐 on AI内参</title>
    <link>https://www.neican.ai/tags/llm%E8%84%91%E8%85%90/</link>
    <description>Recent content in LLM脑腐 on AI内参</description>
    <generator>Hugo</generator>
    <language>zh-cn</language>
    <lastBuildDate>Tue, 21 Oct 2025 12:40:05 +0800</lastBuildDate>
    <atom:link href="https://www.neican.ai/tags/llm%E8%84%91%E8%85%90/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>大模型的“脑腐”之殇：互联网垃圾信息如何侵蚀AI认知，重塑智能未来</title>
      <link>https://www.neican.ai/insights/article-20251021124005080-1/</link>
      <pubDate>Tue, 21 Oct 2025 12:40:05 +0800</pubDate>
      <guid>https://www.neican.ai/insights/article-20251021124005080-1/</guid>
      <description>一项最新研究敲响警钟，大语言模型（LLM）长期接触低质量社交媒体内容将导致类似人类的“认知腐化”，表现为推理、记忆力显著下降，且损伤难以逆转。这揭示了数据质量对AI认知健全的决定性作用，迫使产业界将数据策展提升为AI“认知卫生”的核心，并对未来AGI的构建和数字社会治理提出了严峻挑战。</description>
    </item>
  </channel>
</rss>
