<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>预训练瓶颈 on AI内参</title>
    <link>https://www.neican.ai/tags/%E9%A2%84%E8%AE%AD%E7%BB%83%E7%93%B6%E9%A2%88/</link>
    <description>Recent content in 预训练瓶颈 on AI内参</description>
    <generator>Hugo</generator>
    <language>zh-cn</language>
    <lastBuildDate>Tue, 19 Aug 2025 13:40:04 +0800</lastBuildDate>
    <atom:link href="https://www.neican.ai/tags/%E9%A2%84%E8%AE%AD%E7%BB%83%E7%93%B6%E9%A2%88/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>预训练“拐点”：大模型驶向何方？DeepSeek R2与算力版图的深层困境</title>
      <link>https://www.neican.ai/insights/deepseek-r2-20250819134004602-1/</link>
      <pubDate>Tue, 19 Aug 2025 13:40:04 +0800</pubDate>
      <guid>https://www.neican.ai/insights/deepseek-r2-20250819134004602-1/</guid>
      <description>当前大模型预训练模式已趋近瓶颈，AI的未来在于多模态统一表征和“世界模型”的构建，而非单纯扩大语言模型规模。DeepSeek R2的推迟发布及其与华为昇腾芯片的合作困境，不仅揭示了国产AI算力在性能、成本和生态上的严峻挑战，也预示着全球AI竞赛将进入一个技术范式转移与地缘政治深度交织的新阶段。</description>
    </item>
  </channel>
</rss>
