<?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>GPQA on 硅基 AGI · 智能体学习与测评</title><link>https://guijiagi.com/tags/gpqa/</link><description>Recent content in GPQA on 硅基 AGI · 智能体学习与测评</description><generator>Hugo</generator><language>zh-cn</language><copyright>本站内容采用 CC BY-NC-SA 4.0 国际许可协议授权</copyright><lastBuildDate>Thu, 25 Jun 2026 12:00:00 +0800</lastBuildDate><atom:link href="https://guijiagi.com/tags/gpqa/index.xml" rel="self" type="application/rss+xml"/><item><title>2026 大模型基准测试横评：GPT-5.5 vs Claude 4 vs Gemini 2.5 vs DeepSeek V4</title><link>https://guijiagi.com/posts/llm-benchmark-2026/</link><pubDate>Thu, 25 Jun 2026 12:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/llm-benchmark-2026/</guid><description>全面对比 2026 年主流大语言模型在 MMLU、GPQA、HumanEval、MATH 等基准测试中的表现，附实战性能分析与选型建议。</description></item><item><title>2026 LLM Benchmark 全面解读：MMLU/GPQA/SWE-Bench 谁还有效</title><link>https://guijiagi.com/posts/llm-benchmark-2026-comprehensive/</link><pubDate>Thu, 25 Jun 2026 10:00:00 +0800</pubDate><guid>https://guijiagi.com/posts/llm-benchmark-2026-comprehensive/</guid><description>盘点 2026 年主流 LLM Benchmark 的有效性、数据污染问题与新趋势</description></item></channel></rss>