mlsanityData to 7 Oct 2026

Scaling Test-time Compute for LLM Agents

arXiv 2506.12928 ↗First posted 15 Jun 2025cs.AIcited by 59

On the map: LLM reasoning → Test-time compute and efficient thinking

also in 1 more
Abstract (arXiv)

Scaling test time compute has shown remarkable success in improving the reasoning abilities of large language models (LLMs). In this work, we conduct the first systematic exploration of applying test-time scaling methods to language agents and investigate the extent to which it improves their effectiveness. Specifically, we explore different test-time scaling strategies, including: (1) parallel sampling algorithms; (2) sequential revision strategies; (3) verifiers and merging methods; (4)strategies for diversifying rollouts.We carefully analyze and ablate the impact of different design strategies on applying test-time scaling on language agents, and have follow findings: 1. Scaling test time compute could improve the performance of agents. 2. Knowing when to reflect is important for agents. 3. Among different verification and result merging approaches, the list-wise method performs best. 4. Increasing diversified rollouts exerts a positive effect on the agent's task performance.

Key facts. Cited by 59 papers in the map (data through 7 Oct 2026). Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning. Builds on 25 of its 35 references that are in the map. First posted on arXiv 15 Jun 2025.

Builds on

25 of 35 references.

Closest ancestors · 7

Same direction, newest first

  1. 2025Self-rewarding correction for mathematical reasoning26 Feb
  2. 2025Bag of Tricks for Inference-time Computation of LLM Reasoning11 Feb
  3. 2025Can 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time Scaling10 Feb
  4. 2024A Comparative Study on Reasoning Patterns of OpenAI's o1 Model17 Oct
  5. 2024Training Language Models to Self-Correct via Reinforcement Learning19 Sep
  6. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters6 Aug
  7. 2024Inference Scaling Laws: An Empirical Analysis of Compute-Optimal Inference for Problem-Solving with Language Models1 Aug

Other references · 18

  1. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,146
  2. 2023WebArena: A Realistic Web Environment for Building Autonomous Agentscited by 1,908
  3. 2023MetaGPT: Meta Programming for A Multi-Agent Collaborative Frameworkcited by 1,761
  4. 2024OpenAI o1 System Cardcited by 1,220
  5. 2025Search-o1: Agentic Search-Enhanced Large Reasoning Modelscited by 423
  6. 2025ToolRL: Reward is All Tool Learning Needscited by 342
  7. 2025WebThinker: Empowering Large Reasoning Models with Deep Research Capabilitycited by 320
  8. 2024Magentic-One: A Generalist Multi-Agent System for Solving Complex Taskscited by 221
  9. 2024OS-Copilot: Towards Generalist Computer Agents with Self-Improvementcited by 203
  10. 2025RM-R1: Reward Modeling as Reasoningcited by 149
  11. 2024Tree Search for Language Model Agentscited by 148
  12. 2023Agents: An Open-source Framework for Autonomous Language Agentscited by 116
6 more
  1. 2024Symbolic Learning Enables Self-Evolving Agentscited by 104
  2. 2023RecurrentGPT: Interactive Generation of (Arbitrarily) Long Textcited by 84
  3. 2025AutoAgent: A Fully-Automated and Zero-Code Framework for LLM Agentscited by 49
  4. 2024VisualWebArena: Evaluating Multimodal Agents on Realistic Visual Web Taskscited by 45
  5. 2025Sample, Don't Search: Rethinking Test-Time Alignment for Language Modelscited by 21
  6. 2024TapeAgents: a Holistic Framework for Agent Development and Optimizationcited by 11

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 14; 2026: 45.

59 papers.

Sort:

By area:

  1. 2025ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memorycited by 253
  2. 2025Chain-of-Agents: End-to-End Agent Foundation Models via Multi-Agent Distillation and Agentic RLcited by 75
  3. 2025AgentGym-RL: Training LLM Agents for Long-Horizon Decision Making through Multi-Turn Reinforcement Learningcited by 68
  4. 2025Cognitive Kernel-Pro: A Framework for Deep Research Agents and Agent Foundation Models Trainingcited by 42
  5. 2025Budget-Aware Tool Use Enables Effective Agent Scalingcited by 42
  6. 2025Efficient Agents: Building Effective Agents While Reducing Costcited by 32
  7. 2025Flash-Searcher: Fast and Effective Web Agents via DAG-Based Parallel Executioncited by 27
  8. 2026Large-Scale Terminal Agentic Trajectory Generation from Dockerized Environmentscited by 27
  9. 2026Benchmark Test-Time Scaling of General LLM Agentscited by 24
  10. 2026Agentic Rubrics as Contextual Verifiers for SWE Agentscited by 22
  11. 2026Inference-Time Scaling of Verification: Self-Evolving Deep Research Agents via Test-Time Rubric-Guided Verificationcited by 19
  12. 2025Towards Personalized Deep Research: Benchmarks and Evaluationscited by 14
38 more
  1. 2025Scaling Agents for Computer Usecited by 14
  2. 2025SSRL: Self-Search Reinforcement Learningcited by 13
  3. 2026Inference-Time Budget Control for LLM Search Agentscited by 13
  4. 2026Agent Alpha: Tree Search Unifying Generation, Exploration and Evaluation for Computer-Use Agentscited by 12
  5. 2026When Users Change Their Mind: Evaluating Interruptible Agents in Long-Horizon Web Navigationcited by 12
  6. 2026A Self-Evolving Framework for Efficient Terminal Agents via Observational Context Compressioncited by 12
  7. 2025How Far Are We from Genuinely Useful Deep Research Agents?cited by 11
  8. 2026Scaling Test-Time Compute for Agentic Codingcited by 11
  9. 2026Spend Less, Reason Better: Budget-Aware Value Tree Search for LLM Agentscited by 10
  10. 2026Thinking with Map: Reinforced Parallel Map-Augmented Agent for Geolocalizationcited by 8
  11. 2026The Granularity Mismatch in Agent Security: Argument-Level Provenance Solves Enforcement and Isolates the LLM Reasoning Bottleneckcited by 8
  12. 2026Learning to Ideate for Machine Learning Engineering Agentscited by 6
  13. 2026TSR: Trajectory-Search Rollouts for Multi-Turn RL of LLM Agentscited by 6
  14. 2026Autonomous Research Agents: A Survey of AI Scientists and the Verification Gapcited by 6
  15. 2026Table-as-Search: Formulate Long-Horizon Agentic Information Seeking as Table Completioncited by 5
  16. 2025StyleBench: Evaluating thinking styles in Large Language Modelscited by 4
  17. 2026Search More, Think Less: Rethinking Long-Horizon Agentic Search for Efficiency and Generalizationcited by 4
  18. 2026Same Signal, Opposite Meaning: Direction-Informed Adaptive Learning for LLM Agentscited by 4
  19. 2026IdleSpec: Exploiting Idle Time via Speculative Planning for LLM Agentscited by 4
  20. 2026Poly-EPO: Training Exploratory Reasoning Modelscited by 3
  21. 2026Rethinking Self-Evolving Agent Skills: Feedback Dynamics over Multiple Roundscited by 3
  22. 2026ARTIS: Agentic Risk-Aware Test-Time Scaling via Iterative Simulationcited by 2
  23. 2026CurveShift: Is Agent Progress Scalar? Separating Level from Shapecited by 2
  24. 2026Agentic ESOpt: Fine-Tuning Long-Horizon LLM Agents with Minimal GPU Requirementscited by 2
  25. 2025ARISE: An Adaptive Resolution-Aware Metric for Test-Time Scaling Evaluation in Large Reasoning Modelscited by 1
  26. 2026Self-Pruned Key-Value Attention: Learning When to Write by Predicting Future Utilitycited by 1
  27. 2026Scaling Laws for Agent Harnesses via Effective Feedback Computecited by 1
  28. 2026The Path to Self-Evolving Clinical Systems: Scaling Medical Agents from Assistance to Autonomycited by 1
  29. 2026What is Missing from AI Post-Training AI: An Empirical Analysiscited by 1
  30. 2025ACADREASON: Exploring the Limits of Reasoning Models with Academic Research Problemscited by 0
  31. 2026A Communication-Theoretic Framework for LLM Agents: Cost-Aware Adaptive Reliabilitycited by 0
  32. 2026TClone: Low-Latency Forking of Live GUI Environments for Computer-Use Agentscited by 0
  33. 2026ExComm: Exploration-Stage Communication for Error-Resilient Agentic Test-Time Scalingcited by 0
  34. 2026CIPHER: A Decoupled Exploration-Selection Framework for Test-Time Scaling of Data Science Agentscited by 0
  35. 2026Cognitive Demand Steering for Adaptive Meta-Reasoning in Large Language Modelscited by 0
  36. 2026SkillTV-Bench: Benchmarking How Well Judges Perform on Skill-Augmented Agentic Executioncited by 0
  37. 2026Teach and Grow: An Agent-Centered Architecture for General Robot Learningcited by 0
  38. 2026Disagree to Explore, Agree to Commit: Routing-Guided Test-Time Scaling for Software Agentscited by 0

50 most cited of 59 All 59 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025AgentTTS: Large Language Model Agent for Test-time Compute-optimal Scaling Strategy in Complex Tasks
  2. 2025The Art of Scaling Test-Time Compute for Large Language Models
  3. 2025A Survey on Test-Time Scaling in Large Language Models: What, How, Where, and How Well?
  4. 2025Reasoning on a Budget: A Survey of Adaptive and Controllable Test-Time Compute in LLMs
  5. 2026ThinkBooster: A Unified Framework for Seamless Test-Time Scaling of LLM Reasoning
  6. 2025Test-time Scaling of LLMs: A Survey from A Subproblem Structure Perspective
  7. 2025Are We Scaling the Right Thing? A System Perspective on Test-Time Scaling
  8. 2025Latency and Token-Aware Test-Time Compute