<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>序列模型 on PlumePHP</title><link>https://plumephp.com/tags/%E5%BA%8F%E5%88%97%E6%A8%A1%E5%9E%8B/</link><description>Recent content in 序列模型 on PlumePHP</description><generator>Hugo</generator><language>zh-CN</language><lastBuildDate>Thu, 13 Aug 2026 11:06:00 +0800</lastBuildDate><atom:link href="https://plumephp.com/tags/%E5%BA%8F%E5%88%97%E6%A8%A1%E5%9E%8B/index.xml" rel="self" type="application/rss+xml"/><item><title>06. RNN、LSTM、GRU 与 Transformer</title><link>https://plumephp.com/ai-deep-learning-rnn-transformer/</link><pubDate>Thu, 13 Aug 2026 11:06:00 +0800</pubDate><guid>https://plumephp.com/ai-deep-learning-rnn-transformer/</guid><description>&lt;p&gt;序列数据（文本、时间序列、语音）的建模是深度学习的核心任务之一。从 RNN 的循环结构到 Transformer 的自注意力机制，架构的演进带来了并行化和长距离依赖的突破。本文完整梳理这一技术脉络。&lt;/p&gt;</description></item></channel></rss>