Adaptive Test-Time Compute Allocation with Evolving In-Context Demonstrations

Computer Science > Artificial Intelligence [Submitted on 22 Apr 2026] Adaptive Test-Time Compute Allocation with Evolving In-Context Demonstrations Bowen Zuo, Dongruo Zhou, Yinglun Zhu While scaling test-time compute can substantially improve model performance, existing approaches either rely on static compute allocation or sample from fixed generation distributions. In this work, we introduce a test-time compute allocation framework that jointly adapts where computation is spent and how generation is performed. Our method begins with a warm-up phase that identifies easy queries and assembles an initial pool of question-response pairs from the test set itself. An adaptive phase then concentrates further computation on unresolved queries while reshaping their generation distributions through evolving in-context demonstrations -- conditioning each generation on successful responses from semantically related queries rather than resampling from a fixed distribution. Experiments across math, coding, and reasoning benchmarks demonstrate that our approach consistently outperforms existing baselines while consuming substantially less inference-time compute. Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2604.21018 [cs.AI]   (or arXiv:2604.21018v1 [cs.AI] for this version)   https://doi.org/10.48550/arXiv.2604.21018 Focus to learn more Submission history From: Bowen Zuo [view email] [v1] Wed, 22 Apr 2026 19:07:29 UTC (118 KB) Access Paper: HTML (experimental) view license Current browse context: cs.AI < prev   |   next > new | recent | 2026-04 Change to browse by: cs References & Citations NASA ADS Google Scholar Semantic Scholar Export BibTeX Citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Demos Related Papers About arXivLabs Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)