<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Are LLMs Good Cryptic Crossword Solvers? · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/2403.12094/</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/2403.12094/feed.xml" rel="self" type="application/rss+xml" /><item><title>Are LLMs Good Cryptic Crossword Solvers? — mentioned by nsypteras</title><link>https://news.ycombinator.com/item?id=45929241</link><guid isPermaLink="false">urn:hackerpapers:mention:2403.12094:45929241</guid><pubDate>Fri, 14 Nov 2025 17:35:02 GMT</pubDate><description>&lt;p&gt;Ya interesting thought - would be fascinating if generating games w/solutions is part of the training data pipeline. There&amp;#x27;s been previous work done on on testing LLMs on logic puzzles[1][2][3] so they could possibly be building off those ideas to improve performance. [1] https://huggingface.co/papers/2504.00043 [2] https://huggingface.co/blog/yuchenlin/zebra-logic [3] https://arxiv.org/pdf/2403.12094&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/2403.12094/"&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>
