mlsanityData to 14 Oct 2026

Don't Overthink It: A Survey of Efficient R1-style Large Reasoning Models

arXiv 2508.02120 ↗First posted 4 Aug 2025cs.AIcited by 37

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

Abstract (arXiv)

Recently, Large Reasoning Models (LRMs) have gradually become a research hotspot due to their outstanding performance in handling complex tasks. Among them, DeepSeek R1 has garnered significant attention for its exceptional performance and open-source nature, driving advancements in the research of R1-style LRMs. Unlike traditional Large Language Models (LLMs), these models enhance logical deduction and decision-making capabilities during reasoning by incorporating mechanisms such as long chain-of-thought and self-reflection through reinforcement learning. However, with the widespread application of these models, the problem of overthinking has gradually emerged. Specifically, when generating answers, these models often construct excessively long reasoning chains with redundant or repetitive steps, which leads to reduced reasoning efficiency and may affect the accuracy of the final answer. To this end, various efficient reasoning methods have been proposed, aiming to reduce the length of reasoning paths without compromising model performance and reasoning capability. By reviewing the current research advancements in the field of efficient reasoning methods systematically, we categorize existing works into two main directions based on the lens of single-model optimization versus model collaboration: (1) Efficient Reasoning with Single Model, which focuses on improving the reasoning efficiency of individual models; and (2) Efficient Reasoning with Model Collaboration, which explores optimizing reasoning paths through collaboration among multiple models. Besides, we maintain a public GitHub repository that tracks the latest progress in efficient reasoning methods.

Key facts. Cited by 37 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 164 of its 190 references that are in the map. First posted on arXiv 4 Aug 2025.

Builds on

164 of 190 references.

Closest ancestors · 95

Same direction, newest first

  1. 2025Activation Steering for Chain-of-Thought Compression7 Jul
  2. 2025Controlling Thinking Speed in Reasoning Models4 Jul
  3. 2025AALC: Large Language Model Efficient Reasoning via Adaptive Accuracy-Length Control25 Jun
  4. 2025Fractional Reasoning via Latent Steering Vectors Improves Inference Time Compute18 Jun
  5. 2025Optimizing Length Compression in Large Reasoning Models17 Jun
  6. 2025Efficient Reasoning Through Suppression of Self-Affirmation Reflections in Large Reasoning Models14 Jun
  7. 2025ReCUT: Balancing Reasoning Length and Accuracy in LLMs via Stepwise Trails and Preference Optimization12 Jun
  8. 2025Fast on the Easy, Deep on the Hard: Efficient Reasoning via Powered Length Penalty12 Jun
  9. 2025On Reasoning Strength Planning in Large Reasoning Models10 Jun
  10. 2025Wait, We Don't Need to "Wait"! Removing Thinking Tokens Improves Reasoning Efficiency10 Jun
  11. 2025Not All Tokens Matter: Towards Efficient LLM Reasoning via Token Significance in Reinforcement Learning9 Jun
  12. 2025Overclocking LLM Reasoning: Monitoring and Controlling Thinking Path Lengths in LLMs8 Jun
38 more
  1. 2025How Far Are We from Optimal Reasoning Efficiency?8 Jun
  2. 2025What makes Reasoning Models Different? Follow the Reasoning Leader for Efficient Decoding8 Jun
  3. 2025Route-and-Reason: Scaling Large Language Model Reasoning with Reinforced Model Router6 Jun
  4. 2025SPRINT: Enabling Interleaved Planning and Parallelized Execution in Reasoning Models6 Jun
  5. 2025Just Enough Thinking: Efficient Reasoning with Adaptive Length Penalties Reinforcement Learning5 Jun
  6. 2025OThink-R1: Intrinsic Fast/Slow Thinking Mode Switching for Over-Reasoning Mitigation3 Jun
  7. 2025A*-Thought: Efficient Reasoning via Bidirectional Compression for Low-Resource Settings30 May
  8. 2025AutoL2S: Auto Long-Short Reasoning for Efficient Large Language Models28 May
  9. 2025Mitigating Overthinking in Large Reasoning Models via Manifold Steering28 May
  10. 2025The Price of a Second Thought: On the Evaluation of Reasoning Efficiency in Large Language Models28 May
  11. 2025Don't Think Longer, Think Wisely: Optimizing Thinking Dynamics for Large Reasoning Models27 May
  12. 2025R2R: Efficiently Navigating Divergent Reasoning Paths with Small-Large Model Token Routing27 May
  13. 2025Thinker: Learning to Think Fast and Slow27 May
  14. 2025Self-Route: Automatic Mode Switching via Capability Estimation for Efficient Reasoning27 May
  15. 2025Adaptive Deep Reasoning: Triggering Deep Thinking When Needed26 May
  16. 2025Done Is Better than Perfect: Unlocking Efficient Reasoning by Structured Multi-Turn Decomposition26 May
  17. 2025Interleaved Reasoning for Large Language Models via Reinforcement Learning26 May
  18. 2025Route to Reason: Adaptive Routing for LLM and Reasoning Strategy Selection26 May
  19. 2025LIMOPro: Reasoning Refinement for Efficient and Effective Test-time Scaling25 May
  20. 2025AdaCtrl: Towards Adaptive and Controllable Reasoning via Difficulty-Aware Budgeting24 May
  21. 2025Guided by Gut: Efficient Test-Time Scaling with Reinforced Intrinsic Confidence23 May
  22. 2025Thinking Fast and Right: Balancing Accuracy and Reasoning Length with Adaptive Rewards23 May
  23. 2025Think or Not? Exploring Thinking Efficiency in Large Reasoning Models via an Information-Theoretic Lens23 May
  24. 2025First Finish Search: Efficient Test-Time Scaling in Large Language Models23 May
  25. 2025VeriThinker: Learning to Verify Makes Reasoning Model Efficient23 May
  26. 2025Not All Tokens Are What You Need In Thinking23 May
  27. 2025Don't Overthink it. Preferring Shorter Thinking Chains for Improved LLM Reasoning23 May
  28. 2025TrimR: Verifier-based Training-Free Thinking Compression for Efficient Test-Time Scaling22 May
  29. 2025R1-Compress: Long Chain-of-Thought Compression via Chunk Compression and Search22 May
  30. 2025Incentivizing Dual Process Thinking for Efficient Large Language Model Reasoning22 May
  31. 2025Plan and Budget: Effective and Efficient Test-Time Scaling on Reasoning Large Language Models22 May
  32. 2025Learn to Reason Efficiently with Adaptive Length-based Reward Shaping21 May
  33. 2025Prolonged Reasoning Is Not All You Need: Certainty-Based Adaptive Routing for Efficient LLM/MLLM Reasoning21 May
  34. 2025Think Only When You Need with Large Hybrid-Reasoning Models20 May
  35. 2025Can Pruning Improve Reasoning? Revisiting Long-CoT Compression with Capability in Mind for Better Reasoning20 May
  36. 2025ThinkSwitcher: When to Think Hard, When to Think Fast20 May
  37. 2025FlashThink: An Early Exit Method For Efficient Reasoning20 May
  38. 2025Reasoning Path Compression: Compressing Generation Trajectories for Efficient LLM Reasoning20 May

50 newest of 95 All 95 (JSON)

Other references · 69

  1. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  2. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  3. 2025Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learningcited by 1,444
  4. 2023Representation Engineering: A Top-Down Approach to AI Transparencycited by 1,336
  5. 2024OpenAI o1 System Cardcited by 1,244
  6. 2025Kimi k1.5: Scaling Reinforcement Learning with LLMscited by 1,210
  7. 2022Editing Models with Task Arithmeticcited by 1,120
  8. 2024Refusal in Language Models Is Mediated by a Single Directioncited by 1,032
  9. 2023TIES-Merging: Resolving Interference When Merging Modelscited by 818
  10. 2025Vision-R1: Incentivizing Reasoning Capability in Multimodal Large Language Modelscited by 665
  11. 2025VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Modelcited by 526
  12. 2025Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Modelscited by 384
38 more
  1. 2024Hybrid LLM: Cost-Efficient and Quality-Aware Query Routingcited by 372
  2. 2025R1-Onevision: Advancing Generalized Multimodal Reasoning through Cross-Modal Formalizationcited by 370
  3. 2025ToolRL: Reward is All Tool Learning Needscited by 362
  4. 2025R1-Searcher: Incentivizing the Search Capability in LLMs via Reinforcement Learningcited by 336
  5. 2025From System 1 to System 2: A Survey of Reasoning Large Language Modelscited by 320
  6. 2025Chain of Draft: Thinking Faster by Writing Lesscited by 297
  7. 2025Towards Large Reasoning Models: A Survey of Reinforced Reasoning with Large Language Modelscited by 293
  8. 2024RouterBench: A Benchmark for Multi-LLM Routing Systemcited by 238
  9. 2025Light-R1: Curriculum SFT, DPO and RL for Long COT from Scratch and Beyondcited by 175
  10. 2023Large Language Model Routing with Benchmark Datasetscited by 172
  11. 2024GraphRouter: A Graph-based Router for LLM Selectionscited by 165
  12. 2023Routing to the Expert: Efficient Reward-guided Ensemble of Large Language Modelscited by 145
  13. 2024Cut the Crap: An Economical Communication Pipeline for LLM-based Multi-Agent Systemscited by 141
  14. 2024Mutual Reasoning Makes Smaller LLMs Stronger Problem-Solverscited by 138
  15. 2025Sketch-of-Thought: Efficient LLM Reasoning with Adaptive Cognitive-Inspired Sketchingcited by 117
  16. 2025H-CoT: Hijacking the Chain-of-Thought Safety Reasoning Mechanism to Jailbreak Large Reasoning Models, Including OpenAI o1/o3, DeepSeek-R1, and Gemini 2.0 Flash Thinkingcited by 99
  17. 2025Perception, Reason, Think, and Plan: A Survey on Large Multimodal Reasoning Modelscited by 90
  18. 2025MasRouter: Learning to Route LLMs for Multi-Agent Systemscited by 88
  19. 2025S-GRPO: Early Exit via Reinforcement Learning in Reasoning Modelscited by 87
  20. 2025Visionary-R1: Mitigating Shortcuts in Visual Reasoning with Reinforcement Learningcited by 87
  21. 2025Harnessing Multiple Large Language Models: A Survey on LLM Ensemblecited by 86
  22. 2025Router-R1: Teaching LLMs Multi-Round Routing and Aggregation via Reinforcement Learningcited by 72
  23. 2025Safety in Large Reasoning Models: A Surveycited by 64
  24. 2024Variational Best-of-N Alignmentcited by 51
  25. 2025Stepwise Perplexity-Guided Refinement for Efficient Chain-of-Thought Reasoning in Large Language Modelscited by 51
  26. 2024EmbedLLM: Learning Compact Representations of Large Language Modelscited by 48
  27. 2025IRT-Router: Effective and Interpretable Multi-LLM Routing via Item Response Theorycited by 40
  28. 2025Answer Convergence as a Signal for Early Stopping in Reasoningcited by 40
  29. 2025ARM: Adaptive Reasoning Modelcited by 39
  30. 2025Think When You Need: Self-Adaptive Chain-of-Thought Learningcited by 38
  31. 2025Unlocking General Long Chain-of-Thought Reasoning Capabilities of Large Language Models via Representation Engineeringcited by 36
  32. 2025Search and Refine During Think: Facilitating Knowledge Refinement for Improved Retrieval-Augmented Reasoningcited by 35
  33. 2025Detection and Mitigation of Hallucination in Large Reasoning Models: A Mechanistic Perspectivecited by 34
  34. 2025Ada-R1: Hybrid-CoT via Bi-Level Adaptive Reasoning Optimizationcited by 33
  35. 2025Activation-Informed Merging of Large Language Modelscited by 30
  36. 2025DRP: Distilled Reasoning Pruning with Skill-aware Step Decomposition for Efficient Large Reasoning Modelscited by 30
  37. 2025SafeMLRM: Demystifying Safety in Multi-modal Large Reasoning Modelscited by 29
  38. 2025Adaptive Graph Pruning for Multi-Agent Communicationcited by 29

50 most cited of 69 All 69 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 17; 2026: 20.

37 papers.

Sort:

By area:

  1. 2025Implicit Reasoning in Large Language Models: A Comprehensive Surveycited by 43
  2. 2026VideoAuto-R1: Video Auto Reasoning via Thinking Once, Answering Twicecited by 21
  3. 2025DeceptionBench: A Comprehensive Benchmark for AI Deception Behaviors in Real-world Scenarioscited by 19
  4. 2025ARES: Multimodal Adaptive Reasoning via Difficulty-Aware Token-Level Entropy Shapingcited by 17
  5. 2025DRPO: Efficient Reasoning via Decoupled Reward Policy Optimizationcited by 16
  6. 2026Self-Compression of Chain-of-Thought via Multi-Agent Reinforcement Learningcited by 13
  7. 2025RADAR: Reasoning-Ability and Difficulty-Aware Routing for Reasoning LLMscited by 6
  8. 2025On the Provable Performance Guarantee of Efficient Reasoning Modelscited by 6
  9. 2025DELTA: Dynamic Layer-Aware Token Attention for Efficient Long-Context Reasoningcited by 6
  10. 2026TACT: Mitigating Overthinking and Overacting in Coding Agents via Activation Steeringcited by 6
  11. 2026From Chatbot to Digital Colleague: The Paradigm Shift Toward Persistent Autonomous AIcited by 6
  12. 2025NPG-Muse: Scaling Long Chain-of-Thought Reasoning with NP-Hard Graph Problemscited by 5
25 more
  1. 2025PEAR: Phase Entropy Aware Reward for Efficient Reasoningcited by 5
  2. 2025Explore Briefly, Then Decide: Mitigating LLM Overthinking via Cumulative Entropy Regulationcited by 4
  3. 2025ARM2: Adaptive Reasoning Model with Vision Understanding and Executable Codecited by 4
  4. 2026Capabilities and Fundamental Limits of Latent Chain-of-Thoughtcited by 4
  5. 2026Efficient Reasoning with Balanced Thinkingcited by 4
  6. 2025Towards Self-cognitive Exploration: Metacognitive Knowledge Graph Retrieval Augmented Generationcited by 2
  7. 2025GTAlign: Game-Theoretic Alignment of LLM Assistants for Social Welfarecited by 2
  8. 2025Can Confidence Estimates Decide When Chain-of-Thought Is Necessary for LLMs?cited by 2
  9. 2026Guided by Trajectories: Repairing and Rewarding Tool-Use Trajectories for Tool-Integrated Reasoningcited by 2
  10. 2026Stop Unnecessary Reflection: Training LRMs for Efficient Reasoning with Adaptive Reflection and Length Coordinated Penaltycited by 2
  11. 2026ETR: Entropy Trend Reward for Efficient Chain-of-Thought Reasoningcited by 2
  12. 2026Fewer Tokens, Smaller Cache: Reward-Coordinated Efficient Reasoningcited by 2
  13. 2025EffiReason-Bench: A Unified Benchmark for Evaluating and Advancing Efficient Reasoning in Large Language Modelscited by 1
  14. 2026Fast-Slow Thinking RM: Efficient Integration of Scalar and Generative Reward Modelscited by 1
  15. 2026FlowEvo: Self-Evolving Agents through the Co-Evolution of Workflows and Executable Skillscited by 1
  16. 2025Nav-EE: Navigation-Guided Early Exiting for Efficient Vision-Language Models in Autonomous Drivingcited by 0
  17. 2025ProofSketch: Efficient Verified Reasoning for Large Language Modelscited by 0
  18. 2026Conditional Performance Guarantee for Large Reasoning Modelscited by 0
  19. 2026HiRO-Nav: Hybrid ReasOning Enables Efficient Embodied Navigationcited by 0
  20. 2026Breaking the Reward Barrier: Accelerating Tree-of-Thought Reasoning via Speculative Explorationcited by 0
  21. 2026ThinkRetrieve: Retrieval-Augmented Reasoning Traces for Test-Time Scalingcited by 0
  22. 2026SKIP: a Self-knowledge-guided Step-wise Preference Learning Framework for Concise Reasoningcited by 0
  23. 2026Giving Credit Where It's Due: Redundancy-Aware Learning for Efficient Reasoningcited by 0
  24. 2026FORGE: Form-Optimal Routing of Grounded Evidence for Frozen LLM Agentscited by 0
  25. 2026From Learner Behavior to Reusable Skills for Effective and Efficient Learner Simulationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Towards Concise and Adaptive Thinking in Large Reasoning Models: A Survey
  2. 2025Large Reasoning Models are not thinking straight: on the unreliability of thinking trajectories
  3. 2025A Survey of Reinforcement Learning for Large Reasoning Models
  4. 2025Exploring and Exploiting the Inherent Efficiency within Large Reasoning Models for Self-Guided Efficiency Enhancement
  5. 2025Training Language Models to Reason Efficiently
  6. 2025DeepSeek-R1 Thoughtology: Let's think about LLM Reasoning
  7. 2025An Empirical Study on Eliciting and Improving R1-like Reasoning Models
  8. 2025Trade-offs in Large Reasoning Models: An Empirical Analysis of Deliberative and Adaptive Reasoning over Foundational Capabilities