<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>B'MOJO: Hybrid State Space Realizations of Foundation Models with Eidetic and Fading Memory · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/2407.06324/</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/2407.06324/feed.xml" rel="self" type="application/rss+xml" /><item><title>B'MOJO: Hybrid State Space Realizations of Foundation Models with Eidetic and Fading Memory — mentioned by 100ideas</title><link>https://news.ycombinator.com/item?id=42211207</link><guid isPermaLink="false">urn:hackerpapers:mention:2407.06324:42211207</guid><pubDate>Fri, 22 Nov 2024 04:30:26 GMT</pubDate><description>&lt;p&gt;I think you both make valid points, but I also get the sense that the article is articulating insights gained from pure math explorations into the theoretical limitations of learning, which in the article can sound &amp;quot;turboencabulator-speak&amp;quot; when compressed into words. Maybe I should have just linked to the research paper: [B&amp;#x27;MOJO: Hybrid state space realizations of foundation models with eidetic and fading memory](https://www.arxiv.org/abs/2407.06324)&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/2407.06324/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.LG</category></item></channel></rss>
