<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>What Can Neural Networks Reason About? · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/1905.13211/</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/1905.13211/feed.xml" rel="self" type="application/rss+xml" /><item><title>What Can Neural Networks Reason About? — mentioned by chaxor</title><link>https://news.ycombinator.com/item?id=36065977</link><guid isPermaLink="false">urn:hackerpapers:mention:1905.13211:36065977</guid><pubDate>Thu, 25 May 2023 01:05:15 GMT</pubDate><description>&lt;p&gt;Perhaps, but taken with other works in the area, a better picture does emerge regarding this claim. Substantial works already showing reasoning capabilities in GPT-4, which show that these models do reason very well - near human performance for many causal reasoning tasks. (1) Additionally, there is a mathematical proof that these systems align with dynamic programming, and therefore can perform algorithmic reasoning. (2,3) 1) https://arxiv.org/abs/2305.00050.pdf 2) https://arxiv.org/pdf/1905.13211.pdf 3) https://arxiv.org/pdf/2203.15544.pdf&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/1905.13211/"&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>What Can Neural Networks Reason About? — mentioned by chaxor</title><link>https://news.ycombinator.com/item?id=36065950</link><guid isPermaLink="false">urn:hackerpapers:mention:1905.13211:36065950</guid><pubDate>Thu, 25 May 2023 01:01:18 GMT</pubDate><description>&lt;p&gt;There are substantial works already showing reasoning capabilities in GPT-4, which show that these models do reason extremely well - near human performance for many causal reasoning tasks. (1) Additionally, there is a mathematical proof that these systems align with dynamic programming, and therefore can do algorithmic reasoning. (2,3) 1) https://arxiv.org/abs/2305.00050.pdf 2) https://arxiv.org/pdf/1905.13211.pdf 3) https://arxiv.org/pdf/2203.15544.pdf&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/1905.13211/"&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>
