https://hackerpaper.kaipfister.de/p/2507.13575/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMThttps://news.ycombinator.com/item?id=49749801urn:hackerpapers:mention:2507.13575:49749801Fri, 18 Sep 2026 03:10:40 GMT<p>LLM quants seem to eerily converge to modern/not so modern graphics techniques. You wouldn&#x27;t think it would apply but it&#x27;s obvious in hindsight. In fact mining graphics ideas is probably a good inspiration for efficient LLM architecture. For example, the Hadamard activation transform used here feels a lot like multiplying Fourier basis ala DFT; strong parallels to how image codecs work to make the residuals more compressible (especially discrete block codecs like are used in GPU compressed textures). I thought I was being clever suggesting that you could even abuse texture decode units to…</p><p><a href="https://hackerpaper.kaipfister.de/p/2507.13575/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.LG