<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Article | berumotto-mk</title><link>https://berumotto-mk.github.io/publication_types/article/</link><atom:link href="https://berumotto-mk.github.io/publication_types/article/index.xml" rel="self" type="application/rss+xml"/><description>Article</description><generator>Hugo Blox Builder (https://hugoblox.com)</generator><language>en-us</language><lastBuildDate>Mon, 01 Dec 2025 00:00:00 +0000</lastBuildDate><image><url>https://berumotto-mk.github.io/media/icon_hu7729264130191091259.png</url><title>Article</title><link>https://berumotto-mk.github.io/publication_types/article/</link></image><item><title>Universal Dexterous Functional Grasping via Demonstration-Editing Reinforcement Learning</title><link>https://berumotto-mk.github.io/publication/demofungrasp/</link><pubDate>Mon, 01 Dec 2025 00:00:00 +0000</pubDate><guid>https://berumotto-mk.github.io/publication/demofungrasp/</guid><description>&lt;p>This work introduces DemoFunGrasp, a framework for universal dexterous functional grasping that edits demonstrations to improve sample efficiency and transfer.&lt;/p></description></item></channel></rss>