How Does Code Pretraining Affect Language Model Task Performance? · HackerPapers discussionshttps://hackerpaper.kaipfister.de/p/2409.04556/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMTHow Does Code Pretraining Affect Language Model Task Performance? — mentioned by aspenmartinhttps://news.ycombinator.com/item?id=47542881urn:hackerpapers:mention:2409.04556:47542881Fri, 27 Mar 2026 14:12:49 GMT<p>So what do you think the difference is between humans and an agent in this respect? What makes you think this has any relevance to the problem? everything is combinatorially explosive: the combination of words that we can string into sentences and essays is also combinatorially explosive and yet LLMs and humans have no problem with it. It&#x27;s just the wrong frame of thinking for what&#x27;s going on. These systems are obtaining higher and higher levels of abstractions because that is the most efficient thing for them to do to gain performance. That&#x27;s what reasoning looks like: compositions of higher…</p><p><a href="https://hackerpaper.kaipfister.de/p/2409.04556/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.CL