<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>NeurIPS 2025 on PKU ZhiClass</title><link>https://zhi-class.ai/en/tags/neurips-2025/</link><description>Recent content in NeurIPS 2025 on PKU ZhiClass</description><generator>Hugo -- gohugo.io</generator><language>en</language><copyright>Copyright © Zhi Class 2022-2026</copyright><lastBuildDate>Sun, 01 Jun 2025 00:00:00 +0000</lastBuildDate><atom:link href="https://zhi-class.ai/en/tags/neurips-2025/index.xml" rel="self" type="application/rss+xml"/><item><title>Playing with Transformer at 30+ FPS via Next-Frame Diffusion</title><link>https://zhi-class.ai/en/research/2506.nextframed/</link><pubDate>Sun, 01 Jun 2025 00:00:00 +0000</pubDate><guid>https://zhi-class.ai/en/research/2506.nextframed/</guid><description>In this work, we present Next-Frame Diffusion (NFD), an autoregressive diffusion transformer that incorporates block-wise causal attention, enabling iterative sampling and efficient inference via parallel token generation within each frame.</description><media:content xmlns:media="http://search.yahoo.com/mrss/" url="https://zhi-class.ai/research/2506.nextframed/featured.zh-cn.png"/></item></channel></rss>