https://hackerpaper.kaipfister.de/p/2401.04925/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMThttps://news.ycombinator.com/item?id=39513394urn:hackerpapers:submission:2401.04925:39513394Mon, 26 Feb 2024 16:40:00 GMT<p>The Impact of Reasoning Step Length on Large Language Models</p><p></p><p>2 points; 0 discussion replies on HN. <a href="https://hackerpaper.kaipfister.de/p/2401.04925/">Paper and source history</a>. A submission is not an endorsement.</p>cs.CLhttps://news.ycombinator.com/item?id=39507405urn:hackerpapers:mention:2401.04925:39507405Mon, 26 Feb 2024 03:36:51 GMT<p>I just found this paper i read a while ago. Doesn&#x27;t this answer the question ? The Impact of Reasoning Step Length on Large Language Models - https://arxiv.org/abs/2401.04925 &gt;They discovered that appending dummy tokens (ignored during both training and inference) improves performance somehow. Don’t confuse their guess as to why this might be happening with actual understanding. More tokens is more compute time for the model to utilize, that is completely true. What they guess is that the model can utilize the extra compute for better predictions even if there&#x27;s no extra information to…</p><p><a href="https://hackerpaper.kaipfister.de/p/2401.04925/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.CL