<?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>上下文 on 硅基 AGI · 智能体学习与测评</title><link>https://guijiagi.com/tags/%E4%B8%8A%E4%B8%8B%E6%96%87/</link><description>Recent content in 上下文 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>Sat, 27 Jun 2026 15:00:00 +0800</lastBuildDate><atom:link href="https://guijiagi.com/tags/%E4%B8%8A%E4%B8%8B%E6%96%87/index.xml" rel="self" type="application/rss+xml"/><item><title>多轮对话Prompt优化策略</title><link>https://guijiagi.com/posts/multi-turn-dialogue-optimization/</link><pubDate>Sat, 27 Jun 2026 15:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/multi-turn-dialogue-optimization/</guid><description>多轮对话场景下的Prompt优化策略，解决上下文管理、话题漂移和一致性挑战</description></item><item><title>LLM 上下文长度扩展：从 YARN 到 NTK-aware 插值</title><link>https://guijiagi.com/posts/llm-context-length/</link><pubDate>Thu, 25 Jun 2026 12:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/llm-context-length/</guid><description>系统梳理大语言模型上下文窗口扩展的技术方案，涵盖位置插值、NTK-aware、YaRN、LongRoPE 等方法及其工程实现。</description></item><item><title>上下文窗口扩展技术：从 4K 到 1M</title><link>https://guijiagi.com/posts/context-window-extension/</link><pubDate>Thu, 25 Jun 2026 10:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/context-window-extension/</guid><description>系统梳理 LLM 上下文窗口扩展的核心技术路线与工程实践</description></item></channel></rss>