https://hackerpaper.kaipfister.de/p/2506.05333/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMThttps://news.ycombinator.com/item?id=44203684urn:hackerpapers:mention:2506.05333:44203684Fri, 06 Jun 2025 18:33:22 GMT<p>Hi everyone! We&#x27;re excited to share our new work on rethinking test-time scaling laws. Arxiv link: https://arxiv.org/abs/2506.05333 Website: https://infini-ai-lab.github.io/Kinetics/ Twitter: https://x.com/InfiniAILab/status/1931053042876768586 Summary: (1) Kinetics Scaling Law: We identify a shift in the scaling law for SOTA reasoning models, diverging from existing compute-optimal laws that overlook critical memory access bottlenecks. Our eFLOPs-based holistic analysis reveals that small model effectiveness is often overestimated, suggesting that we should prioritize increasing model size…</p><p><a href="https://hackerpaper.kaipfister.de/p/2506.05333/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.LG