mlsanityData to 14 Oct 2026

Stop Overthinking: A Survey on Efficient Reasoning for Large Language Models

arXiv 2503.16419 ↗First posted 20 Mar 2025cs.CLcited by 464

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

Abstract (arXiv)

Large Language Models (LLMs) have demonstrated remarkable capabilities in complex tasks. Recent advancements in Large Reasoning Models (LRMs), such as OpenAI o1 and DeepSeek-R1, have further improved performance in System-2 reasoning domains like mathematics and programming by harnessing supervised fine-tuning (SFT) and reinforcement learning (RL) techniques to enhance the Chain-of-Thought (CoT) reasoning. However, while longer CoT reasoning sequences improve performance, they also introduce significant computational overhead due to verbose and redundant outputs, known as the "overthinking phenomenon". In this paper, we provide the first structured survey to systematically investigate and explore the current progress toward achieving efficient reasoning in LLMs. Overall, relying on the inherent mechanism of LLMs, we categorize existing works into several key directions: (1) model-based efficient reasoning, which considers optimizing full-length reasoning models into more concise reasoning models or directly training efficient reasoning models; (2) reasoning output-based efficient reasoning, which aims to dynamically reduce reasoning steps and length during inference; (3) input prompts-based efficient reasoning, which seeks to enhance reasoning efficiency based on input prompt properties such as difficulty or length control. Additionally, we introduce the use of efficient data for training reasoning models, explore the reasoning capabilities of small language models, and discuss evaluation methods and benchmarking. Project website: https://github.com/Eclipsess/Awesome-Efficient-Reasoning-LLMs

Key facts. Cited by 464 papers in the map (data through 14 Oct 2026). Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning. Builds on 115 of its 253 references that are in the map. First posted on arXiv 20 Mar 2025.

Builds on

115 of 253 references.

Closest ancestors · 43

Same direction, newest first

  1. 2025InftyThink: Breaking the Length Limits of Long-Context Reasoning in Large Language Models9 Mar
  2. 2025TinyR1-32B-Preview: Boosting Accuracy with Branch-Merge Distillation6 Mar
  3. 2025L1: Controlling How Long A Reasoning Model Thinks With Reinforcement Learning6 Mar
  4. 2025DAST: Difficulty-Adaptive Slow-Thinking for Large Reasoning Models6 Mar
  5. 2025Sampling-Efficient Test-Time Scaling: Self-Estimating the Best-of-N Sampling in Early Decoding3 Mar
  6. 2025How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach3 Mar
  7. 2025Self-Training Elicits Concise Reasoning in Large Language Models27 Feb
  8. 2025Meta-Reasoner: Dynamic Guidance for Optimized Inference-time Reasoning in Large Language Models27 Feb
  9. 2025Efficient Test-Time Scaling via Self-Calibration25 Feb
  10. 2025Towards Thinking-Optimal Scaling of Test-Time Compute for LLM Reasoning25 Feb
  11. 2025Reasoning with Latent Thoughts: On the Power of Looped Transformers24 Feb
  12. 2025Dynamic Parallel Tree Search for Efficient LLM Reasoning22 Feb
31 more
  1. 2025LightThinker: Thinking Step-by-Step Compression21 Feb
  2. 2025Inner Thinking Transformer: Leveraging Dynamic Depth Scaling to Foster Adaptive Internal Thinking19 Feb
  3. 2025Inference-Time Computations for LLM Reasoning and Planning: A Benchmark and Insights18 Feb
  4. 2025TokenSkip: Controllable Chain-of-Thought Compression in LLMs17 Feb
  5. 2025FastMCTS: A Simple Sampling Strategy for Data Synthesis17 Feb
  6. 2025CoT-Valve: Length-Compressible Chain-of-Thought Tuning13 Feb
  7. 2025The Danger of Overthinking: Examining the Reasoning-Action Dilemma in Agentic Tasks12 Feb
  8. 2025Bag of Tricks for Inference-time Computation of LLM Reasoning11 Feb
  9. 2025Can 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time Scaling10 Feb
  10. 2025Confidence Improves Self-Consistency in LLMs10 Feb
  11. 2025Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach7 Feb
  12. 2025Demystifying Long Chain-of-Thought Reasoning in LLMs5 Feb
  13. 2025Token Assorted: Mixing Latent and Text Tokens for Improved Language Model Reasoning5 Feb
  14. 2025Bridging Internal Probability and Self-Consistency for Effective and Efficient LLM Reasoning1 Feb
  15. 2025s1: Simple test-time scaling31 Jan
  16. 2025Reward-Guided Speculative Decoding for Efficient LLM Reasoning31 Jan
  17. 2025Efficient Reasoning with Hidden Thinking31 Jan
  18. 2025Think Smarter not Harder: Adaptive Reasoning with Inference Aware Optimization29 Jan
  19. 2025O1-Pruner: Length-Harmonizing Fine-Tuning for O1-Like Reasoning Pruning22 Jan
  20. 2024Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs30 Dec
  21. 2024Efficiently Scaling LLM Reasoning with Certaindex30 Dec
  22. 2024Token-Budget-Aware LLM Reasoning24 Dec
  23. 2024Compressed Chain of Thought: Efficient Reasoning Through Dense Representations17 Dec
  24. 2024C3oT: Generating Shorter Chain-of-Thought without Compromising Effectiveness16 Dec
  25. 2024Reasoning Aware Self-Consistency: Leveraging Reasoning Paths for Efficient LLM Sampling30 Aug
  26. 2024Path-Consistency with Prefix Enhancement for Efficient Inference in LLMs25 Aug
  27. 2024Make Every Penny Count: Difficulty-Adaptive Self-Consistency for Cost-Efficient Reasoning24 Aug
  28. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters6 Aug
  29. 2024Small Language Models Need Strong Verifiers to Self-Correct Reasoning26 Apr
  30. 2024Escape Sky-high Cost: Early-stopping Self-Consistency for Multi-step Reasoning19 Jan
  31. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Models17 May

Other references · 72

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2017Proximal Policy Optimization Algorithmscited by 15,215
  3. 2024The Llama 3 Herd of Modelscited by 14,715
  4. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  5. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  6. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  7. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  8. 2021Evaluating Large Language Models Trained on Codecited by 8,320
  9. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  10. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,708
  11. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,371
  12. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 5,045
38 more
  1. 2024Qwen2.5 Technical Reportcited by 4,942
  2. 2017Neural Discrete Representation Learningcited by 4,712
  3. 2023Let's Verify Step by Stepcited by 3,789
  4. 2022GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformerscited by 1,919
  5. 2022SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Modelscited by 1,386
  6. 2023AWQ: Activation-aware Weight Quantization for LLM Compression and Accelerationcited by 1,298
  7. 2025Kimi k1.5: Scaling Reinforcement Learning with LLMscited by 1,210
  8. 2024SimPO: Simple Preference Optimization with a Reference-Free Rewardcited by 963
  9. 2022Solving math word problems with process- and outcome-based feedbackcited by 901
  10. 2023Graph of Thoughts: Solving Elaborate Problems with Large Language Modelscited by 886
  11. 2023H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Modelscited by 832
  12. 2024Training Large Language Models to Reason in a Continuous Latent Spacecited by 797
  13. 2024RouteLLM: Learning to Route LLMs with Preference Datacited by 682
  14. 2025Vision-R1: Incentivizing Reasoning Capability in Multimodal Large Language Modelscited by 665
  15. 2024KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cachecited by 637
  16. 2025LIMO: Less is More for Reasoningcited by 498
  17. 2025Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Modelscited by 384
  18. 2025From System 1 to System 2: A Survey of Reasoning Large Language Modelscited by 320
  19. 2025Chain of Draft: Thinking Faster by Writing Lesscited by 297
  20. 2025Towards Large Reasoning Models: A Survey of Reinforced Reasoning with Large Language Modelscited by 293
  21. 2023A Survey on Multimodal Large Language Models for Autonomous Drivingcited by 290
  22. 2021A Survey of Embodied AI: From Simulators to Research Taskscited by 261
  23. 2025Training Language Models to Reason Efficientlycited by 247
  24. 2024From Explicit CoT to Implicit CoT: Learning to Internalize CoT Step by Stepcited by 235
  25. 2025CODI: Compressing Chain-of-Thought into Continuous Space via Self-Distillationcited by 225
  26. 2024Let's Think Dot by Dot: Hidden Computation in Transformer Language Modelscited by 191
  27. 2025When More is Less: Understanding Chain-of-Thought Length in LLMscited by 183
  28. 2025Light-R1: Curriculum SFT, DPO and RL for Long COT from Scratch and Beyondcited by 175
  29. 2023A Survey of Large Language Models for Healthcare: from Data, Technology, and Applications to Accountability and Ethicscited by 163
  30. 2024Distilling System 2 into System 1cited by 159
  31. 2024The Impact of Reasoning Step Length on Large Language Modelscited by 153
  32. 2025Optimizing Test-Time Compute via Meta Reinforcement Fine-Tuningcited by 123
  33. 2025Sketch-of-Thought: Efficient LLM Reasoning with Adaptive Cognitive-Inspired Sketchingcited by 117
  34. 2025SoftCoT: Soft Chain-of-Thought for Efficient Reasoning with LLMscited by 116
  35. 2025Small Models Struggle to Learn from Strong Reasonerscited by 113
  36. 2024Fast Best-of-N Decoding via Speculative Rejectioncited by 112
  37. 2024Mind Your Step (by Step): Chain-of-Thought can Reduce Performance on Tasks where Thinking Makes Humans Worsecited by 109
  38. 2020Mirror Descent Policy Optimizationcited by 106

50 most cited of 72 All 72 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 283; 2026: 181.

464 papers.

Sort:

By area:

  1. 2025OpenThoughts: Data Recipes for Reasoning Modelscited by 304
  2. 2025The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexitycited by 302
  3. 2025Dynamic Early Exit in Reasoning Modelscited by 219
  4. 2025Thinking with Images for Multimodal Reasoning: Foundations, Methods, and Future Frontierscited by 199
  5. 2025ThinkPrune: Pruning Long Chain-of-Thought of LLMs via Reinforcement Learningcited by 192
  6. 2025Reasoning Models Can Be Effective Without Thinkingcited by 187
  7. 2025A Survey of Reinforcement Learning for Large Reasoning Modelscited by 180
  8. 2025Reasoning Models Know When They're Right: Probing Hidden States for Self-Verificationcited by 159
  9. 2025AdaptThink: Reasoning Models Can Learn When to Thinkcited by 147
  10. 2025A Survey of Context Engineering for Large Language Modelscited by 126
  11. 2025Efficient Reasoning Models: A Surveycited by 95
  12. 2025AdaCoT: Pareto-Optimal Adaptive Chain-of-Thought Triggering via Reinforcement Learningcited by 87
38 more
  1. 2025Think Silently, Think Fast: Dynamic Latent Compression of LLM Reasoning Chainscited by 85
  2. 2025Effectively Controlling Reasoning Models through Thinking Interventioncited by 65
  3. 2025Learning When to Think: Shaping Adaptive Reasoning in R1-Style Models via Multi-Stage RLcited by 62
  4. 2025Missing Premise exacerbates Overthinking: Are Reasoning Models losing Critical Thinking Skill?cited by 58
  5. 2025PolyMath: Evaluating Mathematical Reasoning in Multilingual Contextscited by 56
  6. 2025rStar2-Agent: Agentic Reasoning Technical Reportcited by 54
  7. 2025Understanding and Mitigating Numerical Sources of Nondeterminism in LLM Inferencecited by 53
  8. 2025A Survey on Large Language Models for Mathematical Reasoningcited by 52
  9. 2025From Web Search towards Agentic Deep Research: Incentivizing Search with Reasoning Agentscited by 51
  10. 2025Understanding Aha Moments: from External Observations to Internal Mechanismscited by 50
  11. 2025Token Reduction Should Go Beyond Efficiency in Generative Models -- From Vision, Language to Multimodalitycited by 50
  12. 2025AdaCtrl: Towards Adaptive and Controllable Reasoning via Difficulty-Aware Budgetingcited by 50
  13. 2025Inverse Scaling in Test-Time Computecited by 50
  14. 2025Speculative Thinking: Enhancing Small-Model Reasoning with Large Model Guidance at Inference Timecited by 48
  15. 2025SelfBudgeter: Adaptive Token Allocation for Efficient LLM Reasoningcited by 48
  16. 2025Crosslingual Reasoning through Test-Time Scalingcited by 46
  17. 2025SciReplicate-Bench: Benchmarking LLMs in Agent-driven Algorithmic Reproduction from Research Paperscited by 43
  18. 2025Implicit Reasoning in Large Language Models: A Comprehensive Surveycited by 43
  19. 2025Towards Agentic RAG with Deep Reasoning: A Survey of RAG-Reasoning Systems in LLMscited by 42
  20. 2025Answer Convergence as a Signal for Early Stopping in Reasoningcited by 40
  21. 2025Oolong: Evaluating Long Context Reasoning and Aggregation Capabilitiescited by 40
  22. 2025Long-Short Chain-of-Thought Mixture Supervised Fine-Tuning Eliciting Efficient Reasoning in Large Language Modelscited by 39
  23. 2025ARM: Adaptive Reasoning Modelcited by 39
  24. 2025Optimizing Length Compression in Large Reasoning Modelscited by 38
  25. 2025Making Slow Thinking Faster: Compressing LLM Chain-of-Thought via Step Entropycited by 38
  26. 2025ThinkSwitcher: When to Think Hard, When to Think Fastcited by 37
  27. 2025Don't Overthink It: A Survey of Efficient R1-style Large Reasoning Modelscited by 37
  28. 2025Harnessing the Reasoning Economy: A Survey of Efficient Reasoning for Large Language Modelscited by 36
  29. 2025Cost-of-Pass: An Economic Framework for Evaluating Language Modelscited by 35
  30. 2025Prolonged Reasoning Is Not All You Need: Certainty-Based Adaptive Routing for Efficient LLM/MLLM Reasoningcited by 35
  31. 2025Continuous Chain of Thought Enables Parallel Exploration and Reasoningcited by 35
  32. 2025The Homogenizing Effect of Large Language Models on Human Expression and Thoughtcited by 35
  33. 2025OptimalThinkingBench: Evaluating Over and Underthinking in LLMscited by 35
  34. 2025Retro-Search: Exploring Untaken Paths for Deeper and Efficient Reasoningcited by 34
  35. 2025Think or Not? Exploring Thinking Efficiency in Large Reasoning Models via an Information-Theoretic Lenscited by 34
  36. 2025RL Squeezes, SFT Expands: A Comparative Study of Reasoning LLMscited by 34
  37. 2025ConCISE: Confidence-guided Compression in Step-by-step Efficient Reasoningcited by 33
  38. 2025Mitigating Overthinking in Large Reasoning Models via Manifold Steeringcited by 33

50 most cited of 464 All 464 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025Efficient Inference for Large Reasoning Models: A Survey
  2. 2025A Survey of Efficient Reasoning for Large Reasoning Models: Language, Multimodality, and Beyond
  3. 2025Advancing Reasoning in Large Language Models: Promising Methods and Approaches
  4. 2025Logical Reasoning in Large Language Models: A Survey
  5. 2025Large Reasoning Models are not thinking straight: on the unreliability of thinking trajectories
  6. 2026A State-Transition Framework for Efficient LLM Reasoning
  7. 2026Mitigating Overthinking in Large Reasoning Language Models via Reasoning Path Deviation Monitoring
  8. 2026Correct, Concise and Complete: Multi-stage Training For Adaptive Reasoning