<?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>Fine-Tune on 路边的阿不</title>
    <link>https://babyno.top/tags/fine-tune/</link>
    <description>Recent content in Fine-Tune on 路边的阿不</description>
    <generator>Hugo -- gohugo.io</generator>
    <language>zh</language>
    <lastBuildDate>Wed, 14 Jan 2026 09:16:20 +0000</lastBuildDate><atom:link href="https://babyno.top/tags/fine-tune/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>多模态数据提取：微调与少样本提示</title>
      <link>https://babyno.top/posts/2026/01/multimodal-data-extraction-fine-tuning-and-few-shot-prompting/</link>
      <pubDate>Wed, 14 Jan 2026 09:16:20 +0000</pubDate>
      
      <guid>https://babyno.top/posts/2026/01/multimodal-data-extraction-fine-tuning-and-few-shot-prompting/</guid>
      <description>这是一篇偏实践向的记录，主要整理我在「用多模态大模型做发票数据结构化提取」过程中踩过的坑、验证过的方案，以及一些比较稳妥的落地思路。整体目标</description>
    </item>
    
  </channel>
</rss>
