<?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>Virtual-Try-On on MLLog.dev</title><link>https://mllog.dev/en/tags/virtual-try-on/</link><description>Recent content in Virtual-Try-On on MLLog.dev</description><image><title>MLLog.dev</title><url>https://mllog.dev/images/default_mllog.png</url><link>https://mllog.dev/images/default_mllog.png</link></image><generator>Hugo -- 0.147.9</generator><language>en</language><lastBuildDate>Sun, 26 Apr 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://mllog.dev/en/tags/virtual-try-on/index.xml" rel="self" type="application/rss+xml"/><item><title>Tstars-Tryon 1.0: Virtual Try-On as Multi-Image Editing at Taobao Scale</title><link>https://mllog.dev/en/posts/2026-04-26-tstars-tryon-virtual-try-on-mmdit/</link><pubDate>Sun, 26 Apr 2026 00:00:00 +0000</pubDate><guid>https://mllog.dev/en/posts/2026-04-26-tstars-tryon-virtual-try-on-mmdit/</guid><description>How a unified 5B MMDiT trained with multi-reward RL and step distillation reframes virtual try-on as multi-image editing — and runs in under 4 seconds in production.</description></item></channel></rss>