<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>DoFormer | Alireza Karbalayghareh</title><link>https://karbalayghareh.github.io/tag/doformer/</link><atom:link href="https://karbalayghareh.github.io/tag/doformer/index.xml" rel="self" type="application/rss+xml"/><description>DoFormer</description><generator>Wowchemy (https://wowchemy.com)</generator><language>en-us</language><lastBuildDate>Mon, 04 May 2026 00:00:00 +0000</lastBuildDate><image><url>https://karbalayghareh.github.io/media/icon_hua2ec155b4296a9c9791d015323e16eb5_11927_512x512_fill_lanczos_center_2.png</url><title>DoFormer</title><link>https://karbalayghareh.github.io/tag/doformer/</link></image><item><title>DoFormer</title><link>https://karbalayghareh.github.io/project/doformer/</link><pubDate>Mon, 04 May 2026 00:00:00 +0000</pubDate><guid>https://karbalayghareh.github.io/project/doformer/</guid><description>&lt;p>DoFormer is a Transformer for single-cell genomics that encodes gene interventions within attention, distinguishing observational and interventional regimes without requiring a directed acyclic graph. Using biologically informed training objectives, DoFormer improves prediction of held-out gene perturbations relative to baseline methods and transcriptomic foundation models. &lt;a href="https://doi.org/10.64898/2026.05.02.722054" target="_blank" rel="noopener">Preprint on bioRxiv&lt;/a>.&lt;/p></description></item></channel></rss>