A Pretrainer's Guide to Training Data: Measuring the Effects of Data Age, Domain Coverage, Quality, & Toxicity · HackerPapers discussionshttps://hackerpaper.kaipfister.de/p/2305.13169/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMTA Pretrainer's Guide to Training Data: Measuring the Effects of Data Age, Domain Coverage, Quality, & Toxicity — mentioned by light_hue_1https://news.ycombinator.com/item?id=49099065urn:hackerpapers:mention:2305.13169:49099065Wed, 29 Jul 2026 15:50:19 GMT<p>Ironic that &quot;Reinvent science&quot; would publish such a trash article. Did they even read the original paper? This is not at all what it says. https://arxiv.org/pdf/2305.13169 The &quot;poisonous scientific literature&quot; removal is on page 14. Look at the table. Removing academic pieces changes performance by on average 0.44%! You could sneeze and change the performance by that much. You could rerun with a different seed and change the performance by more than that. You could retrain on a different GPU that orders floating point operations differently and change performance by that much. etc. This is…</p><p><a href="https://hackerpaper.kaipfister.de/p/2305.13169/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.CL