mlsanityData to 7 Oct 2026

DynaThink: Fast or Slow? A Dynamic Decision-Making Framework for Large Language Models

arXiv 2407.01009 ↗First posted 1 Jul 2024cs.CLcited by 20

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

Abstract (arXiv)

Large language models (LLMs) have demonstrated emergent capabilities across diverse reasoning tasks via popular Chains-of-Thought (COT) prompting. However, such a simple and fast COT approach often encounters limitations in dealing with complicated problems, while a thorough method, which considers multiple reasoning pathways and verifies each step carefully, results in slower inference. This paper addresses the challenge of enabling LLMs to autonomously select between fast and slow inference methods, thereby optimizing both efficiency and effectiveness. We introduce a dynamic decision-making framework that categorizes tasks into two distinct pathways: 'Fast', designated for tasks where the LLM quickly identifies a high-confidence solution, and 'Slow', allocated for tasks that the LLM perceives as complex and for which it has low confidence in immediate solutions as well as requiring more reasoning paths to verify. Experiments on five popular reasoning benchmarks demonstrated the superiority of the DynaThink over baselines.

Key facts. Cited by 20 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 12 of its 28 references that are in the map. First posted on arXiv 1 Jul 2024.

Builds on

12 of 28 references.

Closest ancestors · 1

Same direction, newest first

  1. 2023SelfCheck: Using LLMs to Zero-Shot Check Their Own Step-by-Step Reasoning1 Aug

Other references · 11

  1. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  2. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,601
  3. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,203
  4. 2022Large Language Models are Zero-Shot Reasonerscited by 4,323
  5. 2023Let's Verify Step by Stepcited by 3,598
  6. 2022Large Language Models Are Human-Level Prompt Engineerscited by 1,020
  7. 2021Are NLP Models really able to Solve Simple Math Word Problems?cited by 861
  8. 2021Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategiescited by 736
  9. 2019MathQA: Towards Interpretable Math Word Problem Solving with Operation-Based Formalismscited by 653
  10. 2017Program Induction by Rationale Generation : Learning to Solve and Explain Algebraic Word Problemscited by 642
  11. 2022Complexity-Based Prompting for Multi-Step Reasoningcited by 387

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 15; 2026: 5.

20 papers.

Sort:

By area:

  1. 2025Between Underthinking and Overthinking: An Empirical Study of Reasoning Length and correctness in LLMscited by 115
  2. 2025AdaCoT: Pareto-Optimal Adaptive Chain-of-Thought Triggering via Reinforcement Learningcited by 86
  3. 2025Think or Not? Exploring Thinking Efficiency in Large Reasoning Models via an Information-Theoretic Lenscited by 32
  4. 2025Reasoning on a Budget: A Survey of Adaptive and Controllable Test-Time Compute in LLMscited by 31
  5. 2025ShadowCoT: Cognitive Hijacking for Stealthy Reasoning Backdoors in LLMscited by 23
  6. 2025Route to Reason: Adaptive Routing for LLM and Reasoning Strategy Selectioncited by 23
  7. 2025Incentivizing Dual Process Thinking for Efficient Large Language Model Reasoningcited by 21
  8. 2025Reasoning on a Spectrum: Aligning LLMs to System 1 and System 2 Thinkingcited by 14
  9. 2025C3PO: Optimized Large Language Model Cascades with Probabilistic Cost Constraints for Reasoningcited by 8
  10. 2025Fast-Slow-Thinking: Complex Task Solving with Large Language Modelscited by 7
  11. 2025On the Provable Performance Guarantee of Efficient Reasoning Modelscited by 6
  12. 2025DynamicMind: A Tri-Mode Thinking System for Large Language Modelscited by 5
8 more
  1. 2026SYMPHONY: Synergistic Multi-agent Planning with Heterogeneous Language Model Assemblycited by 5
  2. 2025Fast, Slow, and Tool-augmented Thinking for LLMs: A Reviewcited by 2
  3. 2025Cognitive Duality for Adaptive Web Agentscited by 1
  4. 2025EconProver: Towards More Economical Test-Time Scaling for Automated Theorem Provingcited by 1
  5. 2026Anytime Safe PAC Efficient Reasoningcited by 1
  6. 2026Towards Open Environments and Instructions: General Vision-Language Navigation via Fast-Slow Interactive Reasoningcited by 0
  7. 2026SafeLens: Deliberate and Efficient Video Guardrails with Fast-and-Slow Screeningcited by 0
  8. 2026EvoThink: Evolving Thinking in Large Reasoning Models via Self-Pruning and Aha-Moment Preference Optimizationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025DynaAct: Large Language Model Reasoning with Dynamic Action Spaces
  2. 2025Stop Overthinking: A Survey on Efficient Reasoning for Large Language Models
  3. 2026Opening the Black Box: A Survey on the Mechanisms of Multi-Step Reasoning in Large Language Models
  4. 2025A Survey of Slow Thinking-based Reasoning LLMs using Reinforced Learning and Inference-time Scaling Law
  5. 2025Advancing Reasoning in Large Language Models: Promising Methods and Approaches
  6. 2025Early Stopping Chain-of-thoughts in Large Language Models
  7. 2025Fast Thinking for Large Language Models
  8. 2023Chain-of-Thought Hub: A Continuous Effort to Measure Large Language Models' Reasoning Performance