<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>CellARC: Measuring Intelligence with Cellular Automata · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/2511.07908/</link><description>Collected HN comments and submissions linking this paper.</description><language>en</language><lastBuildDate>Mon, 02 Nov 2026 11:27:26 GMT</lastBuildDate><atom:link href="https://hackerpaper.kaipfister.de/p/2511.07908/feed.xml" rel="self" type="application/rss+xml" /><item><title>CellARC: Measuring Intelligence with Cellular Automata — HN submission</title><link>https://news.ycombinator.com/item?id=45897072</link><guid isPermaLink="false">urn:hackerpapers:submission:2511.07908:45897072</guid><pubDate>Wed, 12 Nov 2025 06:46:30 GMT</pubDate><description>&lt;p&gt;Show HN: CellARC Measuring Intelligence with Cellular Automata&lt;/p&gt;&lt;p&gt;CellARC, a synthetic benchmark for abstraction and reasoning is built from multicolor 1D cellular automata (CA). Each episode has five support pairs and one query serialized in 256 tokens, enabling rapid iteration with small models while exposing a controllable task space with explicit knobs for alphabet size k, radius r, rule family, Langton&amp;#x27;s lambda, query coverage, and cell entropy. We release 95k training episodes plus two 1k test splits (interpolation/extrapolation) and evaluate symbolic, recurrent, convolutional, transformer, recursive, and LLM baselines. CellARC decouples…&lt;/p&gt;&lt;p&gt;1 point; 0 discussion replies on HN. &lt;a href="https://hackerpaper.kaipfister.de/p/2511.07908/"&gt;Paper and source history&lt;/a&gt;. A submission is not an endorsement.&lt;/p&gt;</description><category>cs.LG</category></item></channel></rss>
