<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Transformers on Orchid Majumder</title><link>https://orchidmajumder.github.io/tags/transformers/</link><description>Recent content in Transformers on Orchid Majumder</description><generator>Hugo</generator><language>en-US</language><lastBuildDate>Sun, 15 Jun 2025 00:00:00 +0000</lastBuildDate><atom:link href="https://orchidmajumder.github.io/tags/transformers/index.xml" rel="self" type="application/rss+xml"/><item><title>Scaling Diffusion Transformers for Text to Image Generation</title><link>https://orchidmajumder.github.io/blog/scaling_diffusion_transformers/</link><pubDate>Sun, 15 Jun 2025 00:00:00 +0000</pubDate><guid>https://orchidmajumder.github.io/blog/scaling_diffusion_transformers/</guid><description>Which diffusion transformer design scales best for text to image generation? UViT does. A 2.3B UViT matches the SDXL UNet, and more data helps every design.</description></item></channel></rss>