Inference-Time Compute Allocation

Overview

This field studies how to distribute limited prediction-time resources—such as reasoning tokens, candidate samples, model calls, search steps, or GPU time—across questions, reasoning paths, and agent actions. Common approaches use difficulty or uncertainty estimates to route tasks, adapt sampling, guide verifier-based search, stop early, or escalate to stronger models; systems work also schedules requests against latency and budget constraints. Evaluations span mathematical reasoning and coding benchmarks as well as retrieval, embodied planning, and multi-tenant inference. A central challenge is that allocation gains depend on whether available signals predict the value of extra computation, and apparent improvements can vanish when costs, selection rules, runtime conditions, and held-out evaluation are controlled.

Latest papers

CardsList

91 papers13 papers in the last four weeks, down 7% on the four weeks before. 0.1% of all new papers.

  1. Sketch, Don't Paint: Allocating Test-Time Compute in Generative Visual Reasoning

    Nov 16, 2026Sheng Lu, Guangpin Tao, Tong Lu +1Inference-Time Compute AllocationTest-Time Scaling for VLMs

  2. Latency-Aware Adaptive Best-of-NN

    Nov 5, 2026Alon Jacobson, Qian Xie, Peter I. FrazierInference-Time Compute AllocationCost-Aware Inference

  3. Give Up to Get More: Priors, Not Policies, Govern Test-Time Compute Allocation

    Nov 4, 2026Jinhyung BaeInference-Time Compute AllocationLLM Inference Scheduling

  4. Certified Access: Test-Time Compute on Conservative Physics-Structured AI Memory

    Oct 29, 2026Pratik Jawahar, Riccardo Maggioni, Jonas Petersen +5Inference-Time Compute AllocationDynamical Systems

  5. What Effort Level Should I Choose? On the Economics of Reasoning Effort in LLM Agents

    Oct 25, 2026Kaiser Sun, Yongchan Kwon, Zain Hasan +3Inference-Time Compute AllocationLLM Agent Efficiency

  6. Markets of Minds: A 51-Billion-State Economy of Intelligence

    Oct 23, 2026Jusheng Zhang, Xuekun Wu, Wenhao Wang +2Multi-Agent Resource AllocationInference-Time Compute Allocation

  7. Where Should a Stuck Agent Spend Its Next Tokens? Causal Attribution of Inference-Time Resources

    Oct 20, 2026Saiyue Lyu, Jerrold Huang, Yunbei Zhang +4Inference-Time Compute AllocationSoftware Engineering Agents

  8. Adaptive Replay Refinement: Allocating Test-Time Compute Between Revision and Resampling

    Oct 20, 2026Ahsan Bilal, Muhammad Ahmed Mohsin, Muhammad Umer +3Inference-Time Compute AllocationTest-Time Scaling

  9. BOSS: Token-Budgeted Online Sampling and Selection for LLM Reasoning

    Oct 19, 2026Yupeng Chang, Yahan Li, Chenlu Guo +3Efficient Language Model ReasoningInference-Time Compute Allocation

  10. Batch-Paced: Fixed-Budget Batch Decoding in Discrete Diffusion Language Models

    Oct 13, 2026Shuai Zhang, Zhanchao Zhou, Huachuan Qiu +2Diffusion Language Model DecodingInference-Time Compute Allocation

  11. Dynamic K-Head: Difficulty-Supervised Budget Selection for Continuous Latent Reasoning

    Oct 13, 2026Ying Li, Yuqi Chen, Sixuan Li +1Efficient Language Model ReasoningInference-Time Compute Allocation

  12. Scaling Inference-Time Compute with Teams of Coding Agents

    Oct 12, 2026Keenan Samway, Daniel D'souza, Omar Darwiche Domingues +3Inference-Time ScalingInference-Time Compute Allocation

  13. Inference-Time Steering Across Scaffolds and Epitopes for Sample-Efficient De Novo Nanobody Design

    Oct 9, 2026Mason Minot, Simon Friedensohn, Cédric R. WeberInference-Time Compute AllocationProtein Binder Design

  14. JevCascade: Readout Stability and Inference-Time Compute Allocation for Prefill-Only Decision Models

    Oct 6, 2026Ran Li, Lei ChenInference-Time Compute AllocationEfficient Language Model Inference

  15. Expanding LLM Reasoning

    Oct 4, 2026Rian Atri, Evan LuoLLM Inference EfficiencyInference-Time Scaling

  16. ReSolve: Reusing Candidate Reasoning through Selective Generative Moderation

    Oct 1, 2026Bangji Yang, Jiajun Fan, Hongbo Ma +6Efficient Language Model ReasoningInference-Time Compute Allocation

  17. The Decision Value of Perception Compute

    Sep 28, 2026Hoang Pham Cong, Ho Viet Duc LuongCost-Aware InferenceAutonomous Driving Perception

  18. ReSync: Re-Aligning the Two Clocks of Asynchronous World-Action Models

    Sep 27, 2026Xi Lin, Feihong Zhang, Yulong Shi +6Action-Conditioned Video GenerationVideo Diffusion Models

  19. SlackDrive: Reclaiming Runtime Slack for Adaptive Driving Inference

    Sep 23, 2026Xiaohuan Pei, Hengguang Zhou, Yuanhao Ban +7Inference-Time Compute AllocationAdaptive Inference

  20. Leveraging Inference-Time Compute for Diffusion Models via Global Scheduling of Denoising Trajectories

    Sep 19, 2026Yuan Cao, Yifu Tang, Hangqi Li +1Inference-Time Compute AllocationDiffusion Sampling