<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>ZeroDL: Zero-shot Distribution Learning for Text Clustering via Large Language Models · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/2406.13342/</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/2406.13342/feed.xml" rel="self" type="application/rss+xml" /><item><title>ZeroDL: Zero-shot Distribution Learning for Text Clustering via Large Language Models — mentioned by jonsmostovojs</title><link>https://news.ycombinator.com/item?id=49643787</link><guid isPermaLink="false">urn:hackerpapers:mention:2406.13342:49643787</guid><pubDate>Thu, 10 Sep 2026 13:56:10 GMT</pubDate><description>&lt;p&gt;Well, you are doing this for free, not me. We are just analysing your data. Actually, as I work on the app with the team, you will be doing even more — namely the brand extraction we subsidise to you currently, will be pushed away to the mac OS app. I am low key fighting internal battles to open source the whole thing, and brand extraction protocol, I believe, will be among the biggest technological contributions. For those interested, we are basing this protocol on a great work by our colleagues at NaverCorp (not an ad!) called &amp;quot;ZeroDL&amp;quot;: https://arxiv.org/abs/2406.13342 I think ultimately…&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/2406.13342/"&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>
