Inference-Time Search

Ask about this topic

Overview

Inference-time search studies how models can use additional computation at deployment to explore alternatives and improve their answers or decisions. Common approaches include best-of-N sampling, beam search and Monte Carlo tree search, guided by verifiers or reward models, with adaptive budgets, pruning, and memory used to focus effort. These methods are applied to mathematical reasoning, code and theorem proving, tool-using agents, planning, and scientific design, where stronger selection can matter as much as generating more candidates. A central tension is that search helps only when its guidance is informative: noisy scores can favor flawed paths, while broader exploration and verification cost time and compute.

Latest papers

CardsList

105 papers14 papers in the last four weeks, up 8% on the four weeks before. 0.1% of all new papers.

  1. SearchHorizonBench: A Benchmark for Search Value in AI Agents

    Oct 26, 2026Jingjie Ning, Xueqi Li, Yibo KongAI Agent BenchmarksInference-Time Search

    Carnegie Mellon University, Pittsburgh, PA, USA

  2. A Systematic Comparison of Search Space Complexity and Efficiency for Time-Synchronous End-to-End ASR Models

    Oct 22, 2026Simon Berger, Larissa Kleppel, Ralf SchlüterInference-Time SearchAutomatic Speech Recognition

    Human Language Technology and Pattern Recognition, Computer Science Department, RWTH Aachen University, Germany · AppTek GmbH, Germany

  3. The Scheduler Is a Hidden Retriever: How Asynchrony Shapes Evidence Acquisition in Search Agents

    Oct 19, 2026Ruiyang Ren, Yuhao Wang, Yinan Zhang +1LLM Inference SchedulingEvidence Retrieval

    Nanyang Technological University Singapore

  4. P(Yes): 15×\times Improvements via Test-Time Scaling of Decision Models

    Oct 18, 2026Lev StamblerLLM Decision-MakingTest-Time Scaling

    Tear Labs · University of Maryland, College Park

  5. Bootstrap Stochastic Interpolant Tree for Inference-time Search Under Sparse Reward

    Oct 16, 2026Binglin Ji, Anindya Sarkar, Yevgeniy VorobeychikInference-Time SearchStochastic Interpolants

    Department of Computer Science & Engineering, Washington University in St. Louis, USA

  6. Uncertainty-Aware Budget Allocation for Efficient LLM Tree Search

    Oct 16, 2026Jianguo Huang, Zhiyi Lyu, Fuxiang Zhang +3Efficient Language Model ReasoningInference-Time Search

    Nanyang Technological University, Singapore

  7. Learning to Search for Executable Patternmaking from Certified Programs

    Oct 15, 2026Zhuotao Xu, Chi Zeng, Liying Yang +2Inference-Time SearchCompiler-Guided Program Synthesis

    Macau University of Science and Technology

  8. Probe-and-Resume: When and How to Evaluate Agent Execution

    Oct 15, 2026Jiakai Xu, Tiancheng Ge, Tujie Guan +4Computer-Use AgentsInference-Time Search

    Columbia University

  9. Input-Conditioned Search for Efficient Test-Time Scaling of Recursive Reasoners

    Oct 13, 2026Haochen Shi, Tianwen Fu, Qianggang Ding +7Inference-Time SearchInference-Time Scaling

    Mila – Quebec AI Institute · Université de Montréal · University of Southern California +2

  10. Resample the Answer: Where to Spend Test-Time Compute in Recursive Reasoning

    Oct 12, 2026Maxime IllouzRecursive Reasoning ModelsInference-Time Search

    Concordia University · Mila – Québec AI Institute · Montréal, Canada

  11. Returning to Development States: When Full Sandbox Snapshots Help Coding Agents

    Oct 9, 2026Daniel Thi Graviet, Tanmay ShahAI Coding AgentsInference-Time Search

    Daytona, San Francisco, California, USA · University of Waterloo, Ontario, Canada

  12. LEVER: Adaptive Cost-Aware Proof Search Over AND/OR Graphs

    Oct 8, 2026Nihal Jain, Shuangjie Yao, Begum Cicekdag +2Automated Theorem ProvingInference-Time Search

    Columbia University

  13. DiMOS: Doob-Guided Inference-Time Multi-Objective Search for Scientific Design

    Oct 5, 2026Ziqing Wang, Qijie Zhu, Weimin Wu +4Masked Diffusion ModelsInference-Time Search

    Northwestern University

  14. Blackboard Intelligence Can Surpass Autoregressive on Globally Constrained Problems

    Sep 30, 2026Woosang Jeon, Jaeyeon Kim, Sham Kakade +6Inference-Time SearchCombinatorial Optimization

    Seoul National University · Harvard University · University of Central Florida

  15. Provable Test-Time Scaling for Beam Search in LLM Reasoning

    Sep 29, 2026Qijia He, Yu Huang, Yuan Cheng +2Long-Horizon LLM ReasoningInference-Time Search

    The Ohio State University · University of Pennsylvania · National University of Singapore

  16. Engineering Efficient Self-Play Chess: Search, Replay, and Throughput Under Limited Compute

    Sep 27, 2026Bertil BraunSelf-Play RLChess

  17. Canopy: Exploiting Piecewise Smooth Tree Priors for Multi-Fidelity Bandits

    Sep 24, 2026Michael Jerge, Suman JanaMulti-Fidelity OptimizationMulti-Armed Bandits

    Amazon · Columbia University

  18. Direct Optimization of Generators for Search in Automated Theorem Proving

    Sep 22, 2026Adam Ousherovitch, Ambuj TewariInference-Time SearchAutomated Theorem Proving

    Department of Statistics University of Michigan

  19. Hill Sampling for Test-Time Scaling: A Simple and Better Alternative to Repeated Sampling, Evolution, and Training

    Sep 22, 2026Jacob Beck, Philip V. Ogren, Ari KobrenInference-Time SearchEvolution Strategies

    Oracle

  20. Self Improvement via Fast Tree-search

    Sep 17, 2026Xinghong Fu, Aravinth Kulanthaivelu, Yutaro YamadaInference-Time SearchAI Coding Agents

    Massachusetts Institute of Technology · Sakana AI

  21. Stellar Colosseum: A Many-Agent Harness for Long-Horizon Research in Mathematics and Theoretical Computer Science

    Sep 14, 2026Honghao Lin, David P. Woodruff, Yuan Deng +3Inference-Time SearchAutomated Theorem Proving

    1Google Research. · 2Carnegie Mellon University.

  22. GraphAHA: Graph-Based Adaptive Search with Heterogeneous Actions for Test-Time Code Generation

    Sep 11, 2026Xitao Li, Haijun Wang, Gege Yuan +4Inference-Time SearchTest-Time Scaling

    Xi’an Jiaotong University · Singapore Management University

  23. Sampling via Decision-Flow: Training-Free Extraction of Improved Latent Reasoning Paths in Large Language Models

    Sep 11, 2026Zhendong Mi, Shaoyi HuangInference-Time SearchLLM Inference

    Stevens Institute of Technology

  24. Rethinking Indirect Prompt Injection as a Test-Time Search Problem

    Sep 3, 2026Duong M. Nguyen, Joon Sik Kim, Blazej Manczak +1Inference-Time SearchAI Agent Security

    Dynamo AI · University of Illinois Urbana-Champaign

  25. Explore Before Committing: Hypothesis-Guided Search for Deep Research Agents

    Sep 1, 2026Ruochen Zhou, Zhengyu Chen, Luan Zhang +3Inference-Time SearchDeep Research Agents

    City University of Hong Kong · Meituan · Independent +1

  26. From Base Rollouts to RL Reasoning: A Budgeted Search Perspective

    Sep 1, 2026Wenhe Sun, Cunxiang Wang, Zijun Yao +1Inference-Time SearchRL for Language Model Reasoning

    1Fudan University · 2Zhipu AI · 3Tsinghua University +1

  27. Escaping Reasoning Basin Collapse with History-Biased Search

    Sep 1, 2026Lu ChengInference-Time SearchReasoning Diversity

    Department of Computer Science and Engineering, Penn State University, University Park, PA 10681

  28. When LLM Meets Tree Search: A Systematic View of Inference as Search in Large Language Models

    Aug 31, 2026Jiaqi Wei, Xiang Zhang, Yuejin Yang +10Inference-Time SearchInference-Time Scaling

    1Zhejiang University · 2Shanghai AI Laboratory · 4Fudan University +3

  29. Mitigating Over-Optimization in PRM-Guided Search in Mathematical Reasoning by Optimizing the Guide

    Aug 30, 2026Taejong Joo, Diego KlabjanReward Over-OptimizationInference-Time Search

    Department of Industrial Engineering & Management Sciences Northwestern University

  30. Operator Packages, Proposer Strength, and Construction-Family Plateaus in Office-Scale Verified Search

    Aug 28, 2026Roberto I. Ono FilhoLLMs for Evolutionary OptimizationLLM-Based Program Synthesis

    Independent researcher