<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>权重合并 on AI内参</title>
    <link>https://www.neican.ai/tags/%E6%9D%83%E9%87%8D%E5%90%88%E5%B9%B6/</link>
    <description>Recent content in 权重合并 on AI内参</description>
    <generator>Hugo</generator>
    <language>zh-cn</language>
    <lastBuildDate>Thu, 03 Jul 2025 22:10:05 +0800</lastBuildDate>
    <atom:link href="https://www.neican.ai/tags/%E6%9D%83%E9%87%8D%E5%90%88%E5%B9%B6/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>大模型“积木式”工程：TNG AoE方法如何以200%加速重塑LLM未来</title>
      <link>https://www.neican.ai/insights/tng-aoe200llm-20250703221005612-0/</link>
      <pubDate>Thu, 03 Jul 2025 22:10:05 +0800</pubDate>
      <guid>https://www.neican.ai/insights/tng-aoe200llm-20250703221005612-0/</guid>
      <description>德国TNG技术咨询公司通过其“专家集合（AoE）”方法，实现了DeepSeek R1-0528模型变体高达200%的推理速度提升。这一突破性技术通过在不重新训练的情况下合并模型权重张量，预示着AI模型开发将从资源密集型训练转向高效的模块化“组装”范式，从而显著降低成本、加速商业应用，并极大地激发开源生态的活力。</description>
    </item>
  </channel>
</rss>
