<?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>文章 on caulif的个人博客</title><link>https://caulif.com/posts/</link><description>Recent content in 文章 on caulif的个人博客</description><generator>Hugo -- 0.154.5</generator><language>zh-cn</language><lastBuildDate>Thu, 18 Jun 2026 00:00:00 +0800</lastBuildDate><atom:link href="https://caulif.com/posts/index.xml" rel="self" type="application/rss+xml"/><item><title>LoRA 学习笔记</title><link>https://caulif.com/posts/lora%E5%AD%A6%E4%B9%A0%E7%AC%94%E8%AE%B0/</link><pubDate>Thu, 18 Jun 2026 00:00:00 +0800</pubDate><guid>https://caulif.com/posts/lora%E5%AD%A6%E4%B9%A0%E7%AC%94%E8%AE%B0/</guid><description>整理 LoRA 的低秩适配原理、训练与推理流程、常见插入位置、核心超参数，以及 QLoRA、AdaLoRA、DoRA 等变体。</description></item><item><title>Agent 学习记录 1.5：Codex 会话记录分析</title><link>https://caulif.com/posts/codex-session-json-trace/</link><pubDate>Fri, 29 May 2026 17:30:00 +0800</pubDate><guid>https://caulif.com/posts/codex-session-json-trace/</guid><description>阅读了一次真实 Codex Desktop 会话的 JSONL 记录，并试图从模型上下文线和 Runtime 事件线两个角度理解 Agent Loop 的运行细节。</description></item><item><title>Agent 学习记录 01：Agent Loop</title><link>https://caulif.com/posts/ai-agent-loop/</link><pubDate>Fri, 29 May 2026 16:30:00 +0800</pubDate><guid>https://caulif.com/posts/ai-agent-loop/</guid><description>从 AI Agent 的基本组成、ReAct 的推理行动循环，以及 Codex 中的 Agent Loop 入手，整理我对 Agent 运行机制的阶段性理解。</description></item><item><title>今天写了一个skill</title><link>https://caulif.com/posts/%E4%BB%8A%E5%A4%A9%E5%86%99%E4%BA%86%E4%B8%80%E4%B8%AAskill/</link><pubDate>Sat, 31 Jan 2026 20:28:07 +0800</pubDate><guid>https://caulif.com/posts/%E4%BB%8A%E5%A4%A9%E5%86%99%E4%BA%86%E4%B8%80%E4%B8%AAskill/</guid><description>记录 page-annotator 的想法、标注交互设计和模型测试过程。</description></item><item><title>GGUF与量化</title><link>https://caulif.com/posts/gguf%E4%B8%8E%E9%87%8F%E5%8C%96/</link><pubDate>Fri, 30 Jan 2026 13:19:41 +0800</pubDate><guid>https://caulif.com/posts/gguf%E4%B8%8E%E9%87%8F%E5%8C%96/</guid><description>从 GGUF 文件结构、内存映射、张量卸载和量化格式入手，整理本地大模型文件与推理优化的基础概念。</description></item><item><title>今日拔了一颗智齿</title><link>https://caulif.com/posts/%E4%BB%8A%E6%97%A5%E6%8B%94%E4%BA%86%E4%B8%80%E9%A2%97%E6%99%BA%E9%BD%BF/</link><pubDate>Thu, 29 Jan 2026 11:57:29 +0800</pubDate><guid>https://caulif.com/posts/%E4%BB%8A%E6%97%A5%E6%8B%94%E4%BA%86%E4%B8%80%E9%A2%97%E6%99%BA%E9%BD%BF/</guid><description>记录一次拔智齿和校医院报销体验。</description></item><item><title>Transformer 学习记录</title><link>https://caulif.com/posts/transfromer%E5%AD%A6%E4%B9%A0%E8%AE%B0%E5%BD%95/</link><pubDate>Wed, 28 Jan 2026 16:34:21 +0800</pubDate><guid>https://caulif.com/posts/transfromer%E5%AD%A6%E4%B9%A0%E8%AE%B0%E5%BD%95/</guid><description>基于 llama2.c 的 run.c 源码注释，梳理 Decoder-only Transformer 的推理流程、RoPE、Attention、FFN、采样和 BPE。</description></item><item><title>博客重新开始</title><link>https://caulif.com/posts/first-post/</link><pubDate>Tue, 27 Jan 2026 17:16:00 +0800</pubDate><guid>https://caulif.com/posts/first-post/</guid><description>很久之前就想弄博客，现在把这里重新整理起来，继续记录技术和生活。</description></item></channel></rss>