Program of Thoughts Prompting: Disentangling Computation from Reasoning for Numerical Reasoning Tasks · HackerPapers discussionshttps://hackerpaper.kaipfister.de/p/2211.12588/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMTProgram of Thoughts Prompting: Disentangling Computation from Reasoning for Numerical Reasoning Tasks — HN submissionhttps://news.ycombinator.com/item?id=46099108urn:hackerpapers:submission:2211.12588:46099108Sun, 30 Nov 2025 18:34:52 GMT<p>Program-of-Thought Prompting Outperforms Chain-of-Thought by 15% (2022)</p><p></p><p>136 points; 36 discussion replies on HN. <a href="https://hackerpaper.kaipfister.de/p/2211.12588/">Paper and source history</a>. A submission is not an endorsement.</p>cs.CLProgram of Thoughts Prompting: Disentangling Computation from Reasoning for Numerical Reasoning Tasks — mentioned by zone411https://news.ycombinator.com/item?id=34333461urn:hackerpapers:mention:2211.12588:34333461Wed, 11 Jan 2023 00:28:23 GMT<p>It&#x27;s possible to enhance mathematical abilities of LLMs by enabling them to externally run symbolic mini programs e.g. https://arxiv.org/abs/2211.12588. It&#x27;s also possible to create fact-grounded retrieval-enhanced language models e.g. https://proceedings.mlr.press/v162/borgeaud22a.html.</p><p><a href="https://hackerpaper.kaipfister.de/p/2211.12588/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.CL