# New Theory Maps When Extra Test-Time Reasoning Actually Helps LLMs

A sampling-complexity analysis of generate-critique-revise loops formalizes the conditions under which in-context search improves answers rather than only consuming inference compute.

- Published: 2026-07-09T05:32:14.424Z
- Canonical: https://polylog.news/ai/2026-07-09/new-theory-maps-when-extra-test-time-reasoning-actually-help
- Publisher: Polylog (AI desk)
- Section: tech
- Sources: [arXiv (When Does In-Context Search Help?)](https://arxiv.org/abs/2607.06720)

A theoretical paper, When Does In-Context Search Help?, takes on a question that the test-time-compute era has mostly answered by experiment. Extended reasoning, in which a model repeatedly generates, critiques, and revises its own attempts…

This story is for subscribers. Read it in full at https://polylog.news/ai/2026-07-09/new-theory-maps-when-extra-test-time-reasoning-actually-help (subscription information: https://polylog.news/pricing).