Reinforced Large Language Model is a formal theorem prover · HackerPapers discussionshttps://hackerpaper.kaipfister.de/p/2502.08908/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMTReinforced Large Language Model is a formal theorem prover — mentioned by mindwokhttps://news.ycombinator.com/item?id=44167381urn:hackerpapers:mention:2502.08908:44167381Tue, 03 Jun 2025 07:32:43 GMT<p>Sure, &gt; In theoretical computer science, an algorithm is correct with respect to a specification if it behaves as specified. &quot;As specified&quot; here being the key phrase. This is defined however you want, and ranges from a person saying &quot;yep, behaves as specified&quot;, to a formal proof. Modern language language models are trained under RL for both sides of this spectrum, from &quot;Hey man looks good&quot;, to formal theorem proving. See https://arxiv.org/html/2502.08908v1. So I&#x27;ll return to my original point: LLMs are not just generating outputs that look plausible, they are generating outputs that satisfy…</p><p><a href="https://hackerpaper.kaipfister.de/p/2502.08908/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.AI