<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Peng Cheng | 厦门大学语言智能技术课题组 XMU-DeepLIT Deep Language Intelligence Technology of Xiamen University</title><link>https://xmudeeplit.github.io/author/peng-cheng/</link><atom:link href="https://xmudeeplit.github.io/author/peng-cheng/index.xml" rel="self" type="application/rss+xml"/><description>Peng Cheng</description><generator>Hugo Blox Builder (https://hugoblox.com)</generator><language>zh-Hans</language><lastBuildDate>Sat, 22 Aug 2026 09:36:36 +0000</lastBuildDate><image><url>https://xmudeeplit.github.io/media/logo_hu_d39049b37e1fa974.png</url><title>Peng Cheng</title><link>https://xmudeeplit.github.io/author/peng-cheng/</link></image><item><title>Training Matryoshka Mixture-of-Experts for Elastic Inference-Time Expert Utilization</title><link>https://xmudeeplit.github.io/publication/196/</link><pubDate>Sat, 22 Aug 2026 09:36:36 +0000</pubDate><guid>https://xmudeeplit.github.io/publication/196/</guid><description/></item></channel></rss>