<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Decoding Game: On Minimax Optimality of Heuristic Text Generation Strategies · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/2410.03968/</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/2410.03968/feed.xml" rel="self" type="application/rss+xml" /><item><title>Decoding Game: On Minimax Optimality of Heuristic Text Generation Strategies — mentioned by Der_Einzige</title><link>https://news.ycombinator.com/item?id=43890210</link><guid isPermaLink="false">urn:hackerpapers:mention:2410.03968:43890210</guid><pubDate>Sun, 04 May 2025 22:38:32 GMT</pubDate><description>&lt;p&gt;Besides beam search and it&amp;#x27;s variants? (there are many including the little known but awesomely powerful constrained beam search: https://huggingface.co/blog/constrained-beam-search) Does MBR (minimal bayes risk) sampling count? Also there was this paper at ICLR which is relevant to this question: https://arxiv.org/abs/2410.03968 This paper basically claims that non-heuristic methods (like beam search) are harmful compared to the heuristic ones.&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/2410.03968/"&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>
