Cost-Effective Agent Harnesses for Abstract Reasoning and Generalization on ARC-AGI-1 · HackerPapers discussionshttps://hackerpaper.kaipfister.de/p/2607.06764/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMTCost-Effective Agent Harnesses for Abstract Reasoning and Generalization on ARC-AGI-1 — mentioned by Legend2440https://news.ycombinator.com/item?id=48948617urn:hackerpapers:mention:2607.06764:48948617Fri, 17 Jul 2026 15:32:56 GMT<p>No, you are misunderstanding the paper. https://arxiv.org/abs/2607.06764 The base model got 15%. They built an elaborate looping harness that allows it to burn 100k tokens &quot;thinking&quot; about the problem, which got the 57%. This is just an alternative approach to reasoning.</p><p><a href="https://hackerpaper.kaipfister.de/p/2607.06764/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.AI