<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Can LLMs plan paths in the real world? · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/2411.17912/</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/2411.17912/feed.xml" rel="self" type="application/rss+xml" /><item><title>Can LLMs plan paths in the real world? — mentioned by nathanfig</title><link>https://news.ycombinator.com/item?id=50047360</link><guid isPermaLink="false">urn:hackerpapers:mention:2411.17912:50047360</guid><pubDate>Sun, 11 Oct 2026 20:15:08 GMT</pubDate><description>&lt;p&gt;For anyone interested in digging deeper on LLM navigation capabilities, there is the work I mentioned in the post[0], and also a paper that tried to test this more formally in 2024[1]. I didn&amp;#x27;t learn of these until after I did my experiment, it would have been interesting to retest the same routes. Maybe in a year, when everything has changed once again. [0] https://github.com/Hansenq/nav-evals-public [1] https://arxiv.org/abs/2411.17912&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/2411.17912/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.AI</category></item></channel></rss>
