<?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>Reflexion on 硅基 AGI · 智能体学习与测评</title><link>https://guijiagi.com/tags/reflexion/</link><description>Recent content in Reflexion 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 10:21:00 +0800</lastBuildDate><atom:link href="https://guijiagi.com/tags/reflexion/index.xml" rel="self" type="application/rss+xml"/><item><title>AI Agent的规划能力：从ReAct到Tree-of-Planning</title><link>https://guijiagi.com/posts/b2-af6834ab/</link><pubDate>Thu, 16 Jul 2026 10:21:00 +0800</pubDate><guid>https://guijiagi.com/posts/b2-af6834ab/</guid><description>系统梳理AI Agent规划算法的发展脉络，分析ReAct、Reflexion、LATS等规划方法的原理与适用场景</description></item><item><title>从ReAct到Reflexion：Agent推理范式演进</title><link>https://guijiagi.com/posts/article-38/</link><pubDate>Sun, 12 Jul 2026 23:10:00 +0800</pubDate><guid>https://guijiagi.com/posts/article-38/</guid><description>Agent推理范式的三次跃迁：从ReAct到Reflexion再到自主规划的演进逻辑</description></item><item><title>从ReAct到Reflexion：Agent推理范式演进</title><link>https://guijiagi.com/posts/b2-64fc97fc/</link><pubDate>Sun, 12 Jul 2026 23:10:00 +0800</pubDate><guid>https://guijiagi.com/posts/b2-64fc97fc/</guid><description>Agent推理范式的三次跃迁：从ReAct到Reflexion再到自主规划的演进逻辑</description></item><item><title>Agent反思与自我纠错架构：让AI学会从错误中学习</title><link>https://guijiagi.com/posts/agent-reflection-self-correction/</link><pubDate>Thu, 02 Jul 2026 10:05:00 +0800</pubDate><guid>https://guijiagi.com/posts/agent-reflection-self-correction/</guid><description>深入探讨Agent反思机制的设计原理、自我纠错架构与实践模式</description></item></channel></rss>