<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>数学推理 on Pluto</title><link>https://uestcyz.top/tags/%E6%95%B0%E5%AD%A6%E6%8E%A8%E7%90%86/</link><description>Recent content in 数学推理 on Pluto</description><image><title>Pluto</title><url>https://uestcyz.top/social.png</url><link>https://uestcyz.top/social.png</link></image><generator>Hugo</generator><language>zh-CN</language><lastBuildDate>Sat, 31 May 2025 21:23:55 +0000</lastBuildDate><atom:link href="https://uestcyz.top/tags/%E6%95%B0%E5%AD%A6%E6%8E%A8%E7%90%86/index.xml" rel="self" type="application/rss+xml"/><item><title>论文阅读：SuperCorrect</title><link>https://uestcyz.top/posts/research-supercorrect/</link><pubDate>Sat, 31 May 2025 21:23:55 +0000</pubDate><guid>https://uestcyz.top/posts/research-supercorrect/</guid><description>SuperCorrect 论文阅读：两阶段微调，借助大模型的层次化思维模板与跨模型协作 DPO，提升小模型的数学推理与自我纠错能力。</description></item><item><title>论文阅读：ReasonFlux</title><link>https://uestcyz.top/posts/research-reasonflux/</link><pubDate>Sun, 11 May 2025 04:35:40 +0000</pubDate><guid>https://uestcyz.top/posts/research-reasonflux/</guid><description>ReasonFlux 论文阅读：构建思维模板库，并通过分层强化学习规划模板轨迹，提升 LLM 的复杂数学推理能力。</description></item></channel></rss>