<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>AI基础架构 on AI内参</title>
    <link>https://www.neican.ai/tags/ai%E5%9F%BA%E7%A1%80%E6%9E%B6%E6%9E%84/</link>
    <description>Recent content in AI基础架构 on AI内参</description>
    <generator>Hugo</generator>
    <language>zh-cn</language>
    <lastBuildDate>Tue, 15 Jul 2025 12:10:05 +0800</lastBuildDate>
    <atom:link href="https://www.neican.ai/tags/ai%E5%9F%BA%E7%A1%80%E6%9E%B6%E6%9E%84/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>突破Transformer桎梏：EBT“能量驱动”架构如何开启通用推理新纪元？</title>
      <link>https://www.neican.ai/insights/transformerebt-20250715121005799-1/</link>
      <pubDate>Tue, 15 Jul 2025 12:10:05 +0800</pubDate>
      <guid>https://www.neican.ai/insights/transformerebt-20250715121005799-1/</guid>
      <description>能量驱动Transformer（EBT）架构通过引入能量建模，颠覆了传统AI的前馈推理模式，实现了动态计算、不确定度建模和自我验证三大突破，模拟人类System 2思维，大幅提升了训练效率、泛化能力和多模态应用潜力。这项创新不仅挑战了Transformer的霸主地位，更描绘了通向通用人工智能和下一代智能计算范式的宏伟蓝图。</description>
    </item>
  </channel>
</rss>
