https://hackerpaper.kaipfister.de/p/2510.14901/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMThttps://news.ycombinator.com/item?id=46255969urn:hackerpapers:submission:2510.14901:46255969Sat, 13 Dec 2025 16:57:44 GMT<p>Reasoning with Sampling: Your Base Model Is Smarter Than You Think</p><p></p><p>2 points; 0 discussion replies on HN. <a href="https://hackerpaper.kaipfister.de/p/2510.14901/">Paper and source history</a>. A submission is not an endorsement.</p>cs.LGhttps://news.ycombinator.com/item?id=46145119urn:hackerpapers:mention:2510.14901:46145119Thu, 04 Dec 2025 08:25:52 GMT<p>The warmup process is necessary in order to try to find high-probability regions of the target distribution. That&#x27;s not an issue for an LLM, since it&#x27;s trained to sample directly from a distribution which looks like natural language. There is some work on using MCMC to sample from higher-probability regions of an LLM distribution [1], but that&#x27;s a separate thing. Nobody doubts that an LLM is sampling from its target distribution from the first token it outputs. [1] https://arxiv.org/abs/2510.14901</p><p><a href="https://hackerpaper.kaipfister.de/p/2510.14901/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.LGhttps://news.ycombinator.com/item?id=45749234urn:hackerpapers:submission:2510.14901:45749234Wed, 29 Oct 2025 16:32:51 GMT<p>Reasoning with Sampling: Your Base Model Is Smarter Than You Think</p><p></p><p>2 points; 0 discussion replies on HN. <a href="https://hackerpaper.kaipfister.de/p/2510.14901/">Paper and source history</a>. A submission is not an endorsement.</p>cs.LGhttps://news.ycombinator.com/item?id=45717457urn:hackerpapers:submission:2510.14901:45717457Mon, 27 Oct 2025 04:39:56 GMT<p>Reasoning with Sampling: Your Base Model Is Smarter Than You Think</p><p></p><p>1 point; 0 discussion replies on HN. <a href="https://hackerpaper.kaipfister.de/p/2510.14901/">Paper and source history</a>. A submission is not an endorsement.</p>cs.LGhttps://news.ycombinator.com/item?id=45654328urn:hackerpapers:submission:2510.14901:45654328Tue, 21 Oct 2025 10:36:23 GMT<p>Reasoning with Sampling: Your Base Model Is Smarter Than You Think</p><p></p><p>3 points; 0 discussion replies on HN. <a href="https://hackerpaper.kaipfister.de/p/2510.14901/">Paper and source history</a>. A submission is not an endorsement.</p>cs.LGhttps://news.ycombinator.com/item?id=45646167urn:hackerpapers:submission:2510.14901:45646167Mon, 20 Oct 2025 16:55:24 GMT<p>Reasoning with Sampling: Your Base Model Is Smarter Than You Think</p><p></p><p>2 points; 0 discussion replies on HN. <a href="https://hackerpaper.kaipfister.de/p/2510.14901/">Paper and source history</a>. A submission is not an endorsement.</p>cs.LG