<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Unsupervised Cross-Domain Image Generation · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/1611.02200/</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/1611.02200/feed.xml" rel="self" type="application/rss+xml" /><item><title>Unsupervised Cross-Domain Image Generation — mentioned by axegon_</title><link>https://news.ycombinator.com/item?id=24439484</link><guid isPermaLink="false">urn:hackerpapers:mention:1611.02200:24439484</guid><pubDate>Fri, 11 Sep 2020 05:34:05 GMT</pubDate><description>&lt;p&gt;For years we&amp;#x27;ve been hearing how AI will build robots and wipe out humanity and the bollocks that is the trolley problem. I remember when I was reading the Unsupervised Cross-Domain Image Generation[1] paper and my immediate thought was &amp;quot;yep, I can see this going south&amp;quot;. And sure enough, not long after deepfakes became a thing. GPT-3 is absolutely astonishing in terms of it&amp;#x27;s capabilities and I&amp;#x27;d love to be able to dig into it&amp;#x27;s inner workings and scroll through it&amp;#x27;s code. The truth is there are three stoppers for the large majority of people who would love to exploit it. 1. Data. For better…&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/1611.02200/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.CV</category></item></channel></rss>
