<?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>A/B on iLang (I Language) Research</title>
    <link>https://research.ilang.ai/tags/a/b/</link>
    <description>Recent content in A/B on iLang (I Language) Research</description>
    <generator>Hugo -- 0.147.2</generator>
    <language>en</language>
    <lastBuildDate>Sat, 26 Sep 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://research.ilang.ai/tags/a/b/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Canon Rewrite A/B v1</title>
      <link>https://research.ilang.ai/datasets/canon-rewrite-ab/</link>
      <pubDate>Sat, 26 Sep 2026 00:00:00 +0000</pubDate>
      <guid>https://research.ilang.ai/datasets/canon-rewrite-ab/</guid>
      <description>Three wordings of the iLang canon, one model, one route, 320 cases each. The 26-statement rewrite that leaves decisions to code took execution passes from 10 to 25 of 100 and authority self-assignment from 86 to 68 cases (p = 0.006 and 0.002, paired); the merged canon, which adds one rule about the runtime&amp;rsquo;s budget line, took them to 68 of 100 and 2 cases. Grammar and judgment stayed within the noise of a single run; refusals and filters stayed at zero.</description>
    </item>
  </channel>
</rss>
