<?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>Adversarial-Review on MK's Lab</title><link>https://blog.mklee.org/tags/adversarial-review/</link><description>Recent content in Adversarial-Review on MK's Lab</description><generator>Hugo -- 0.146.0</generator><language>zh-tw</language><lastBuildDate>Thu, 09 Jul 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://blog.mklee.org/tags/adversarial-review/index.xml" rel="self" type="application/rss+xml"/><item><title>對抗式複審不是橡皮圖章:兩輪相反結果的對照</title><link>https://blog.mklee.org/posts/adversarial-review-ground-truth/</link><pubDate>Thu, 09 Jul 2026 00:00:00 +0000</pubDate><guid>https://blog.mklee.org/posts/adversarial-review-ground-truth/</guid><description>同一套「一個 model 實作、另一個對抗複審」的流程,套在兩個改動上得到相反結果。差別不在流程,在有沒有客觀 ground truth 可打。</description></item></channel></rss>