<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Limits of Deep Learning: Sequence Modeling through the Lens of Complexity Theory · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/2405.16674/</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/2405.16674/feed.xml" rel="self" type="application/rss+xml" /><item><title>Limits of Deep Learning: Sequence Modeling through the Lens of Complexity Theory — mentioned by namaria</title><link>https://news.ycombinator.com/item?id=43756065</link><guid isPermaLink="false">urn:hackerpapers:mention:2405.16674:43756065</guid><pubDate>Mon, 21 Apr 2025 20:23:37 GMT</pubDate><description>&lt;p&gt;Theoretical limitations of multi-layer Transformer https://arxiv.org/abs/2412.02975 On Limitations of the Transformer Architecture https://arxiv.org/abs/2402.08164 Limits of Deep Learning: Sequence Modeling through the Lens of Complexity Theory https://arxiv.org/abs/2405.16674 TL;DR transformers are inherently limited with tasks requiring composition of sequential steps&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/2405.16674/"&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>
