<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>The Universal Weight Subspace Hypothesis · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/2512.05117/</link><description>Collected HN comments and submissions linking this paper.</description><language>en</language><lastBuildDate>Mon, 02 Nov 2026 11:16:22 GMT</lastBuildDate><atom:link href="https://hackerpaper.kaipfister.de/p/2512.05117/feed.xml" rel="self" type="application/rss+xml" /><item><title>The Universal Weight Subspace Hypothesis — mentioned by refibrillator</title><link>https://news.ycombinator.com/item?id=49760576</link><guid isPermaLink="false">urn:hackerpapers:mention:2512.05117:49760576</guid><pubDate>Fri, 18 Sep 2026 21:34:49 GMT</pubDate><description>&lt;p&gt;You might like The Universal Weight Subspace Hypothesis: https://arxiv.org/abs/2512.05117 Curious to know if anyone is aware of research trying what parent suggested?&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/2512.05117/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.LG</category></item><item><title>The Universal Weight Subspace Hypothesis — mentioned by thoughtpeddler</title><link>https://news.ycombinator.com/item?id=48338360</link><guid isPermaLink="false">urn:hackerpapers:mention:2512.05117:48338360</guid><pubDate>Sat, 30 May 2026 16:57:54 GMT</pubDate><description>&lt;p&gt;Ya, super interesting research area the authors explored of basically trying to answer the question: &amp;quot;Is there a canonical/intrinsic way that concepts/representations/information are &amp;#x27;stored&amp;#x27; in the universe/reality?&amp;quot;. They tested that by performing &amp;quot;spectral analysis of over 1100 models - including 500 Mistral-7B LoRAs, 500 Vision Transformers, and 50 LLaMA-8B models ... by applying spectral decomposition techniques to the weight matrices of various architectures&amp;quot;, and concluding that &amp;quot;deep neural networks trained across diverse tasks exhibit remarkably similar low-dimensional parametric…&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/2512.05117/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.LG</category></item><item><title>The Universal Weight Subspace Hypothesis — HN submission</title><link>https://news.ycombinator.com/item?id=46199623</link><guid isPermaLink="false">urn:hackerpapers:submission:2512.05117:46199623</guid><pubDate>Tue, 09 Dec 2025 00:16:46 GMT</pubDate><description>&lt;p&gt;The universal weight subspace hypothesis&lt;/p&gt;&lt;p&gt;&lt;/p&gt;&lt;p&gt;358 points; 132 discussion replies on HN. &lt;a href="https://hackerpaper.kaipfister.de/p/2512.05117/"&gt;Paper and source history&lt;/a&gt;. A submission is not an endorsement.&lt;/p&gt;</description><category>cs.LG</category></item><item><title>The Universal Weight Subspace Hypothesis — HN submission</title><link>https://news.ycombinator.com/item?id=46193683</link><guid isPermaLink="false">urn:hackerpapers:submission:2512.05117:46193683</guid><pubDate>Mon, 08 Dec 2025 15:50:17 GMT</pubDate><description>&lt;p&gt;The Universal Weight Subspace Hypothesis&lt;/p&gt;&lt;p&gt;&lt;/p&gt;&lt;p&gt;1 point; 1 discussion reply on HN. &lt;a href="https://hackerpaper.kaipfister.de/p/2512.05117/"&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>
