<?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>Co&#39;Log</title>
    <link>https://coldeye2020.github.io/</link>
    <description>Recent content on Co&#39;Log</description>
    <generator>Hugo -- 0.139.4</generator>
    <language>en</language>
    <lastBuildDate>Sun, 19 Apr 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://coldeye2020.github.io/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>深入理解 Python GIL：从原理到实践</title>
      <link>https://coldeye2020.github.io/tech/2026-04-19-python-gil/</link>
      <pubDate>Sun, 19 Apr 2026 00:00:00 +0000</pubDate>
      <guid>https://coldeye2020.github.io/tech/2026-04-19-python-gil/</guid>
      <description>从 CPython 引用计数机制出发，通过实验数据揭示 GIL 对 CPU 密集型任务的拖累，并为机器学习、深度学习、量化金融从业者提供系统性的规避策略。</description>
    </item>
    <item>
      <title>让 AI 自己学会挖掘股票信号？聊聊 QuantaAlpha 这篇论文</title>
      <link>https://coldeye2020.github.io/invest/2026-04-11-quantaalpha-popular/</link>
      <pubDate>Sat, 11 Apr 2026 14:00:00 +0800</pubDate>
      <guid>https://coldeye2020.github.io/invest/2026-04-11-quantaalpha-popular/</guid>
      <description>介绍 QuantaAlpha 论文：一个让 LLM 像量化研究员一样自主挖掘、进化 Alpha 因子的框架，解析其核心机制与实验结果。</description>
    </item>
    <item>
      <title>从&#34;绝活哥&#34;到认知负荷控制</title>
      <link>https://coldeye2020.github.io/think/2026-04-11-cognitive-load/</link>
      <pubDate>Sat, 11 Apr 2026 00:00:00 +0800</pubDate>
      <guid>https://coldeye2020.github.io/think/2026-04-11-cognitive-load/</guid>
      <description>从英雄联盟&amp;#34;绝活哥&amp;#34;现象出发，结合高中备考经历，探讨主动降低认知负荷如何让人腾出思考空间、实现跨越式进步。</description>
    </item>
    <item>
      <title>OmniDrive 论文解读</title>
      <link>https://coldeye2020.github.io/tech/2024-12-15-omnidrive/</link>
      <pubDate>Sun, 15 Dec 2024 11:42:16 +0800</pubDate>
      <guid>https://coldeye2020.github.io/tech/2024-12-15-omnidrive/</guid>
      <description>解读 OmniDrive：一种将多模态大语言模型引入自动驾驶的 3D 场景理解框架，介绍其 Q-Former3D 结构设计与反事实推断评测方式。</description>
    </item>
    <item>
      <title>The begining</title>
      <link>https://coldeye2020.github.io/think/2024-12-15-introduction/</link>
      <pubDate>Sun, 15 Dec 2024 01:39:39 +0800</pubDate>
      <guid>https://coldeye2020.github.io/think/2024-12-15-introduction/</guid>
      <description>博客的开篇，记录 Coldeye 与 OuhaoTen 建立这个博客的初心——对抗拖延、留存记录，以及对未来持续更新的期待。</description>
    </item>
  </channel>
</rss>
