<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Silent Tokens, Loud Effects: Padding in LLMs · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/2510.01238/</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/2510.01238/feed.xml" rel="self" type="application/rss+xml" /><item><title>Silent Tokens, Loud Effects: Padding in LLMs — mentioned by dist-epoch</title><link>https://news.ycombinator.com/item?id=49993958</link><guid isPermaLink="false">urn:hackerpapers:mention:2510.01238:49993958</guid><pubDate>Wed, 07 Oct 2026 15:12:48 GMT</pubDate><description>&lt;p&gt;Since you&amp;#x27;re already setup, try adding padding to the input prompt. Add 1000 dots &amp;quot;.&amp;quot; after the question, this provides extra space where to reason about the problem. It&amp;#x27;s not as good as real reasoning, since it&amp;#x27;s not recursive, but it does improve the quality. https://arxiv.org/abs/2510.01238&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/2510.01238/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.CL</category></item></channel></rss>
