<?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>模型训练 on 硅基 AGI · 智能体学习与测评</title><link>https://guijiagi.com/tags/%E6%A8%A1%E5%9E%8B%E8%AE%AD%E7%BB%83/</link><description>Recent content in 模型训练 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>Wed, 08 Jul 2026 13:00:00 +0800</lastBuildDate><atom:link href="https://guijiagi.com/tags/%E6%A8%A1%E5%9E%8B%E8%AE%AD%E7%BB%83/index.xml" rel="self" type="application/rss+xml"/><item><title>Hermes 4微调实战：从数据准备到模型部署全流程</title><link>https://guijiagi.com/posts/hermes4-finetuning-guide/</link><pubDate>Wed, 08 Jul 2026 13:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/hermes4-finetuning-guide/</guid><description>Nous Hermes 4微调完整指南：数据准备、LoRA训练、评估调优、量化部署的企业级实践</description></item><item><title>指令微调配方详解：打造高质量监督微调数据集</title><link>https://guijiagi.com/posts/instruction-tuning-recipe/</link><pubDate>Thu, 02 Jul 2026 10:45:00 +0800</pubDate><guid>https://guijiagi.com/posts/instruction-tuning-recipe/</guid><description>深入探讨LLM指令微调的数据构建、配方设计与训练最佳实践</description></item><item><title>LoRA vs QLoRA 2026对比：参数高效微调的两种路径</title><link>https://guijiagi.com/posts/lora-vs-qlora-2026/</link><pubDate>Thu, 02 Jul 2026 10:42:00 +0800</pubDate><guid>https://guijiagi.com/posts/lora-vs-qlora-2026/</guid><description>系统对比LoRA与QLoRA在2026年的技术差异、性能表现与适用场景</description></item><item><title>Self-RAG实现指南：让模型自己决定何时检索</title><link>https://guijiagi.com/posts/self-rag-implementation/</link><pubDate>Thu, 02 Jul 2026 10:41:00 +0800</pubDate><guid>https://guijiagi.com/posts/self-rag-implementation/</guid><description>深入探讨Self-RAG的原理、实现方法与生产实践，让模型自主控制检索行为</description></item></channel></rss>