<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>COMET: Commonsense Transformers for Automatic Knowledge Graph Construction · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/1906.05317/</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/1906.05317/feed.xml" rel="self" type="application/rss+xml" /><item><title>COMET: Commonsense Transformers for Automatic Knowledge Graph Construction — mentioned by MacsHeadroom</title><link>https://news.ycombinator.com/item?id=35352214</link><guid isPermaLink="false">urn:hackerpapers:mention:1906.05317:35352214</guid><pubDate>Wed, 29 Mar 2023 03:52:18 GMT</pubDate><description>&lt;p&gt;I don&amp;#x27;t understand what you want me to ask it about. So I asked it something, as usual every citation and URL is real despite not having internet access or training data after 2021: &amp;lt;PROMPT&amp;gt;List some citations of recent Machine Learning papers discussing reasoning ability in Machine Learning Models such as HellaSwag. [REPLY]&amp;quot; As of my knowledge cutoff in September 2021, I can provide you with some citations for ML papers that discuss reasoning abilities in models and datasets like HellaSwag. However, you may want to search for more recent papers using a platform like arXiv or Google Scholar.…&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/1906.05317/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.CL</category></item><item><title>COMET: Commonsense Transformers for Automatic Knowledge Graph Construction — mentioned by wfhpw</title><link>https://news.ycombinator.com/item?id=25647734</link><guid isPermaLink="false">urn:hackerpapers:mention:1906.05317:25647734</guid><pubDate>Tue, 05 Jan 2021 17:10:41 GMT</pubDate><description>&lt;p&gt;Has there been any good work applying causal inference explicitly to common NLP tasks like question-answering? Or to game-play? The lack of common-sense in models like GPT-3 would seem to indicate a huge opportunity, perhaps in tandem with COMET[0] or some other graph-based approach. [0] https://arxiv.org/abs/1906.05317&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/1906.05317/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.CL</category></item><item><title>COMET: Commonsense Transformers for Automatic Knowledge Graph Construction — mentioned by p1esk</title><link>https://news.ycombinator.com/item?id=23595833</link><guid isPermaLink="false">urn:hackerpapers:mention:1906.05317:23595833</guid><pubDate>Sun, 21 Jun 2020 22:17:26 GMT</pubDate><description>&lt;p&gt;You are kidding, right? All these questions you provided are extremely simple to answer, compared to many other things clever human interrogators might say during TT. I&amp;#x27;m starting to doubt your NLP expertise. The TT ready model I&amp;#x27;m envisioning will be trained on many billions of chat sessions. It will contain dozens of preconstructed graphs and will dynamically construct dozens more (personality graph, common sense knowledge graph, domain specific knowledge graphs, causality graph, dialog state graph, emotional state graph, etc), it will have a bunch of emotion detectors, humor detectors…&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/1906.05317/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.CL</category></item><item><title>COMET: Commonsense Transformers for Automatic Knowledge Graph Construction — mentioned by p1esk</title><link>https://news.ycombinator.com/item?id=23365372</link><guid isPermaLink="false">urn:hackerpapers:mention:1906.05317:23365372</guid><pubDate>Sat, 30 May 2020 22:28:26 GMT</pubDate><description>&lt;p&gt;I think using AE for text generation is a good idea, and is pretty old one (I tried it myself back in 2015 without particularly good results), but I wouldn&amp;#x27;t call it a breakthrough. To me a breakthrough/new paradigm would be something like this: https://arxiv.org/abs/1906.05317&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/1906.05317/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.CL</category></item><item><title>COMET: Commonsense Transformers for Automatic Knowledge Graph Construction — HN submission</title><link>https://news.ycombinator.com/item?id=20195954</link><guid isPermaLink="false">urn:hackerpapers:submission:1906.05317:20195954</guid><pubDate>Sun, 16 Jun 2019 15:33:13 GMT</pubDate><description>&lt;p&gt;Comet: Commonsense Transformers for Automatic Knowledge Graph Construction&lt;/p&gt;&lt;p&gt;&lt;/p&gt;&lt;p&gt;4 points; 0 discussion replies on HN. &lt;a href="https://hackerpaper.kaipfister.de/p/1906.05317/"&gt;Paper and source history&lt;/a&gt;. A submission is not an endorsement.&lt;/p&gt;</description><category>cs.CL</category></item></channel></rss>
