<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>深度学习 on 南巷</title><link>https://nx.wmlab.top/tags/%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0/</link><description>Recent content in 深度学习 on 南巷</description><generator>Hugo -- gohugo.io</generator><language>zh-CN</language><managingEditor>liuwenhao1968@163.com (南巷)</managingEditor><webMaster>liuwenhao1968@163.com (南巷)</webMaster><copyright>© 2026 南巷</copyright><lastBuildDate>Sun, 14 Jun 2026 19:06:46 +0800</lastBuildDate><atom:link href="https://nx.wmlab.top/tags/%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0/index.xml" rel="self" type="application/rss+xml"/><item><title>Flow Matching</title><link>https://nx.wmlab.top/blog/flow-matching/</link><pubDate>Sun, 14 Jun 2026 19:06:46 +0800</pubDate><author>liuwenhao1968@163.com (南巷)</author><guid>https://nx.wmlab.top/blog/flow-matching/</guid><description>从 Diffusion 的加噪、去噪与监督目标出发，逐步理解 Flow Matching 如何学习从噪声到数据的连续流场。</description></item><item><title>ACoT-VLA</title><link>https://nx.wmlab.top/blog/acot-vla/</link><pubDate>Sun, 14 Jun 2026 19:06:05 +0800</pubDate><author>liuwenhao1968@163.com (南巷)</author><guid>https://nx.wmlab.top/blog/acot-vla/</guid><description>围绕 ACoT-VLA 的背景、模型架构、显式/隐式动作推理与动作引导预测，梳理 VLA 中语义理解到低层动作控制的过渡设计。</description></item><item><title>LLaMA 2 SFT 监督微调</title><link>https://nx.wmlab.top/blog/llama2-sft-supervised-finetuning/</link><pubDate>Sun, 14 Jun 2026 17:52:56 +0800</pubDate><author>liuwenhao1968@163.com (南巷)</author><guid>https://nx.wmlab.top/blog/llama2-sft-supervised-finetuning/</guid><description>系统整理 SFT 监督微调的目标、数据格式、训练方式与代码流程，帮助模型从通用能力走向任务对齐。</description></item><item><title>LLaMA 2 预训练流程</title><link>https://nx.wmlab.top/blog/llama2-pretraining-process/</link><pubDate>Sun, 14 Jun 2026 17:52:46 +0800</pubDate><author>liuwenhao1968@163.com (南巷)</author><guid>https://nx.wmlab.top/blog/llama2-pretraining-process/</guid><description>从数据集构建、样本切分、词表处理、模型输入到训练循环，梳理 LLaMA 2 预训练的核心流程。</description></item><item><title>LLaMA 2 动手实现</title><link>https://nx.wmlab.top/blog/llama2-hands-on-implementation/</link><pubDate>Sun, 14 Jun 2026 17:52:34 +0800</pubDate><author>liuwenhao1968@163.com (南巷)</author><guid>https://nx.wmlab.top/blog/llama2-hands-on-implementation/</guid><description>围绕 LLaMA 2 的数据集、Tokenizer、模型结构与训练代码，记录从原理拆解到工程实现的完整实践过程。</description></item><item><title>LLaMA 2 解析</title><link>https://nx.wmlab.top/blog/llama2-analysis/</link><pubDate>Sun, 14 Jun 2026 17:49:58 +0800</pubDate><author>liuwenhao1968@163.com (南巷)</author><guid>https://nx.wmlab.top/blog/llama2-analysis/</guid><description>从模型架构、Tokenizer、RMSNorm、GQA、RoPE、FFN 到输出层，系统拆解 LLaMA 2 的核心组件与设计动机。</description></item></channel></rss>