https://hackerpaper.kaipfister.de/p/2110.14168/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMThttps://news.ycombinator.com/item?id=37476213urn:hackerpapers:mention:2110.14168:37476213Tue, 12 Sep 2023 02:01:33 GMT<p>During the big GPT-4 news cycle I think a bunch of folks posted claims that were outrageously good- &quot;language model passes medical exams better than humans&quot;, etc. When I looked into them, in nearly all cases, the claims were boosted far beyond the reality. And the reality seemed much more consistent with a fairly banal interpretation: LLMs produce realistic looking text but have no real ability to distinguish truth from fabrication (which is a step beyond bullshit!). The one example that still interests me is math problem solving. Can next-token predictors really solve generalized math…</p><p><a href="https://hackerpaper.kaipfister.de/p/2110.14168/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.LG