<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Editing Models with Task Arithmetic · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/2212.04089/</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/2212.04089/feed.xml" rel="self" type="application/rss+xml" /><item><title>Editing Models with Task Arithmetic — HN submission</title><link>https://news.ycombinator.com/item?id=39895907</link><guid isPermaLink="false">urn:hackerpapers:submission:2212.04089:39895907</guid><pubDate>Mon, 01 Apr 2024 16:26:28 GMT</pubDate><description>&lt;p&gt;Editing Models with Task Arithmetic (2022)&lt;/p&gt;&lt;p&gt;&lt;/p&gt;&lt;p&gt;1 point; 0 discussion replies on HN. &lt;a href="https://hackerpaper.kaipfister.de/p/2212.04089/"&gt;Paper and source history&lt;/a&gt;. A submission is not an endorsement.&lt;/p&gt;</description><category>cs.LG</category></item><item><title>Editing Models with Task Arithmetic — HN submission</title><link>https://news.ycombinator.com/item?id=35388856</link><guid isPermaLink="false">urn:hackerpapers:submission:2212.04089:35388856</guid><pubDate>Fri, 31 Mar 2023 15:45:41 GMT</pubDate><description>&lt;p&gt;Editing Models with Task Arithmetic&lt;/p&gt;&lt;p&gt;&lt;/p&gt;&lt;p&gt;3 points; 0 discussion replies on HN. &lt;a href="https://hackerpaper.kaipfister.de/p/2212.04089/"&gt;Paper and source history&lt;/a&gt;. A submission is not an endorsement.&lt;/p&gt;</description><category>cs.LG</category></item><item><title>Editing Models with Task Arithmetic — mentioned by convexfunction</title><link>https://news.ycombinator.com/item?id=34680761</link><guid isPermaLink="false">urn:hackerpapers:mention:2212.04089:34680761</guid><pubDate>Mon, 06 Feb 2023 17:24:38 GMT</pubDate><description>&lt;p&gt;It&amp;#x27;s not quite the same, and might not be possible to make rigorous enough that it really proves anything, but something sort of similar would actually be practical to at least attempt in many cases. Stable Diffusion checkpoints of the same major version, along with other families of model weights, have the IMO fascinating property that you can do element-wise arithmetic with them, and the resulting model will actually sort of function like you&amp;#x27;d naively expect. Recent paper on the topic (in LLMs, not diffusion models) here: https://arxiv.org/abs/2212.04089 So, if you take a Stable Diffusion…&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/2212.04089/"&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>
