<?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>Generative-Models on Quang Minh Dinh</title>
    <link>https://quangminhdinh.github.io/tags/generative-models/</link>
    <description>Recent content in Generative-Models on Quang Minh Dinh</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    <lastBuildDate>Sat, 20 Jun 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://quangminhdinh.github.io/tags/generative-models/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>The loss decides what the model ignores</title>
      <link>https://quangminhdinh.github.io/garden/the-loss-decides-what-the-model-ignores/</link>
      <pubDate>Sat, 20 Jun 2026 00:00:00 +0000</pubDate>
      <guid>https://quangminhdinh.github.io/garden/the-loss-decides-what-the-model-ignores/</guid>
      <description>Two projects where the objective, not the architecture, dictated what the model paid attention to and what it left blank.</description>
    </item>
    <item>
      <title>Every conditioning choice is a hidden assumption</title>
      <link>https://quangminhdinh.github.io/garden/conditioning-is-an-assumption/</link>
      <pubDate>Thu, 15 Jan 2026 00:00:00 +0000</pubDate>
      <guid>https://quangminhdinh.github.io/garden/conditioning-is-an-assumption/</guid>
      <description>Lessons from extending IMLE to text-conditional generation: failures almost always traced to latent mismatch.</description>
    </item>
  </channel>
</rss>
