<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Vision-Language on 硅基 AGI · 智能体学习与测评</title><link>https://guijiagi.com/tags/vision-language/</link><description>Recent content in Vision-Language on 硅基 AGI · 智能体学习与测评</description><generator>Hugo</generator><language>zh-cn</language><copyright>本站内容采用 CC BY-NC-SA 4.0 国际许可协议授权</copyright><lastBuildDate>Thu, 16 Jul 2026 11:09:00 +0800</lastBuildDate><atom:link href="https://guijiagi.com/tags/vision-language/index.xml" rel="self" type="application/rss+xml"/><item><title>多模态大模型技术原理：从CLIP到原生统一架构</title><link>https://guijiagi.com/posts/b1-4a19c4f9/</link><pubDate>Thu, 16 Jul 2026 11:09:00 +0800</pubDate><guid>https://guijiagi.com/posts/b1-4a19c4f9/</guid><description>系统讲解多模态大模型的技术演进，从CLIP双塔到GPT-4V统一架构再到原生多模态训练</description></item></channel></rss>