arxiv
PublishedJune 30, 2026 at 4:00 AM
—neutral
When More Sampling Hurts: The Modal Ceiling and Correlation Ceiling of Test-Time Scaling
Publisher summary· verbatim
arXiv:2606.28661v1 Announce Type: cross Abstract: People overthink; language models over-sample, and the extra effort can talk both into a worse answer. Reasoning systems answer a hard question by sampling it many times (test-time scaling), and the more they draw, the more often a correct answer tur
Stay posted· Newsletter
A 5-min weekly brief — top movers, price watch, story of the week.
Discussion
No replies yet. Be first.
Related coverage
More from ARXIV
arxivDiffusion-warm sampling of the XY model enables fast thermalization at scale8harxivFrom Idea to Prototype in an Afternoon: Scaffolded, AI-Assisted Rapid VA Prototyping8harxiv3D HAMSTER: Bridging Planning and Control in Hierarchical Vision Language Action Models through 3D Trajectory Guidance8harxivImproving LLM Reasoning with Homophily-aware Structural and Semantic Text-Attributed Graph Compression8hThe Bubble Brief
WEEKLYRead AI insights every Tuesday — top movers, new releases, story of the week.
Originally published on arxiv ↗