mlsanityData to 7 Oct 2026

O1-Pruner: Length-Harmonizing Fine-Tuning for O1-Like Reasoning Pruning

arXiv 2501.12570 ↗First posted 22 Jan 2025cs.CLcited by 278

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

Abstract (arXiv)

Recently, long-thought reasoning LLMs, such as OpenAI's O1, adopt extended reasoning processes similar to how humans ponder over complex problems. This reasoning paradigm significantly enhances the model's problem-solving abilities and has achieved promising results. However, long-thought reasoning process leads to a substantial increase in inference time. A pressing challenge is reducing the inference overhead of long-thought LLMs while ensuring accuracy. In this paper, we experimentally demonstrate that long-thought reasoning models struggle to effectively allocate token budgets based on problem difficulty and reasoning redundancies. To address this, we propose Length-Harmonizing Fine-Tuning (O1-Pruner), aiming at minimizing reasoning overhead while maintaining accuracy. This effective fine-tuning method first estimates the LLM's baseline performance through pre-sampling and then uses RL-style fine-tuning to encourage the model to generate shorter reasoning processes under accuracy constraints. This allows the model to achieve efficient reasoning with lower redundancy while maintaining accuracy. Experiments on various mathematical reasoning benchmarks show that O1-Pruner not only significantly reduces inference overhead but also achieves higher accuracy, providing a novel and promising solution to this challenge. Our code is coming soon at https://github.com/StarDewXXX/O1-Pruner

Key facts. Cited by 278 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 21 of its 28 references that are in the map. First posted on arXiv 22 Jan 2025.

Builds on

21 of 28 references.

Closest ancestors · 6

Same direction, newest first

  1. 2024Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs30 Dec
  2. 2024Token-Budget-Aware LLM Reasoning24 Dec
  3. 2024Compressed Chain of Thought: Efficient Reasoning Through Dense Representations17 Dec
  4. 2024C3oT: Generating Shorter Chain-of-Thought without Compromising Effectiveness16 Dec
  5. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters6 Aug
  6. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Models17 May

Other references · 15

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  2. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  3. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 8,851
  4. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 7,916
  5. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,767
  6. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,203
  7. 2019The Curious Case of Neural Text Degenerationcited by 2,881
  8. 2024SimPO: Simple Preference Optimization with a Reference-Free Rewardcited by 951
  9. 2023Graph of Thoughts: Solving Elaborate Problems with Large Language Modelscited by 873
  10. 2024Training Large Language Models to Reason in a Continuous Latent Spacecited by 763
  11. 2024Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Surveycited by 641
  12. 2022Towards Reasoning in Large Language Models: A Surveycited by 593
3 more
  1. 2023Large Language Model Alignment: A Surveycited by 241
  2. 2023Evaluating the Performance of Large Language Models on GAOKAO Benchmarkcited by 182
  3. 2024Mulberry: Empowering MLLM with o1-like Reasoning and Reflection via Collective Monte Carlo Tree Searchcited by 171

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 178; 2026: 100.

278 papers.

Sort:

By area:

  1. 2025Stop Overthinking: A Survey on Efficient Reasoning for Large Language Modelscited by 456
  2. 2025Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Modelscited by 381
  3. 2025From System 1 to System 2: A Survey of Reasoning Large Language Modelscited by 319
  4. 2025DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoningcited by 261
  5. 2025Training Language Models to Reason Efficientlycited by 242
  6. 2025Dynamic Early Exit in Reasoning Modelscited by 217
  7. 2025Deep Think with Confidencecited by 199
  8. 2025When More is Less: Understanding Chain-of-Thought Length in LLMscited by 182
  9. 2025Reasoning Models Can Be Effective Without Thinkingcited by 182
  10. 2025A Survey of Reinforcement Learning for Large Reasoning Modelscited by 175
  11. 2026GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimizationcited by 175
  12. 2025CoT-Valve: Length-Compressible Chain-of-Thought Tuningcited by 167
38 more
  1. 2025DAST: Difficulty-Adaptive Slow-Thinking for Large Reasoning Modelscited by 154
  2. 2025A Survey of Efficient Reasoning for Large Reasoning Models: Language, Multimodality, and Beyondcited by 131
  3. 2025Safety Tax: Safety Alignment Makes Your Large Reasoning Models Less Reasonablecited by 120
  4. 2025Towards Thinking-Optimal Scaling of Test-Time Compute for LLM Reasoningcited by 118
  5. 2025Between Underthinking and Overthinking: An Empirical Study of Reasoning Length and correctness in LLMscited by 115
  6. 2025Thinkless: LLM Learns When to Thinkcited by 107
  7. 2025MedReason: Eliciting Factual Medical Reasoning Steps in LLMs via Knowledge Graphscited by 105
  8. 2025LightThinker: Thinking Step-by-Step Compressioncited by 104
  9. 2025Self-Training Elicits Concise Reasoning in Large Language Modelscited by 97
  10. 2025Efficient Reasoning Models: A Surveycited by 93
  11. 2025S-GRPO: Early Exit via Reinforcement Learning in Reasoning Modelscited by 87
  12. 2025AdaCoT: Pareto-Optimal Adaptive Chain-of-Thought Triggering via Reinforcement Learningcited by 86
  13. 2025ShorterBetter: Guiding Reasoning Models to Find Optimal Inference Length for Efficient Reasoningcited by 70
  14. 2025Revisiting the Test-Time Scaling of o1-like Models: Do they Truly Possess Test-Time Scaling Capabilities?cited by 69
  15. 2025Quantization Hurts Reasoning? An Empirical Study on Quantized Reasoning Modelscited by 68
  16. 2025Wait, We Don't Need to "Wait"! Removing Thinking Tokens Improves Reasoning Efficiencycited by 65
  17. 2025Think Only When You Need with Large Hybrid-Reasoning Modelscited by 64
  18. 2025Pangu Embedded: An Efficient Dual-system LLM Reasoner with Metacognitioncited by 61
  19. 2025Learning When to Think: Shaping Adaptive Reasoning in R1-Style Models via Multi-Stage RLcited by 59
  20. 2025Unlocking Efficient Long-to-Short LLM Reasoning with Model Mergingcited by 58
  21. 2025Efficient Inference for Large Reasoning Models: A Surveycited by 58
  22. 2025Missing Premise exacerbates Overthinking: Are Reasoning Models losing Critical Thinking Skill?cited by 58
  23. 2025Hume: Introducing System-2 Thinking in Visual-Language-Action Modelcited by 56
  24. 2025R1-ShareVL: Incentivizing Reasoning Capability of Multimodal Large Language Models via Share-GRPOcited by 52
  25. 2025AdaCtrl: Towards Adaptive and Controllable Reasoning via Difficulty-Aware Budgetingcited by 50
  26. 2025Token Reduction Should Go Beyond Efficiency in Generative Models -- From Vision, Language to Multimodalitycited by 49
  27. 2025Inverse Scaling in Test-Time Computecited by 49
  28. 2025SelfBudgeter: Adaptive Token Allocation for Efficient LLM Reasoningcited by 47
  29. 2025Answer Convergence as a Signal for Early Stopping in Reasoningcited by 38
  30. 2025Think When You Need: Self-Adaptive Chain-of-Thought Learningcited by 37
  31. 2025Long-Short Chain-of-Thought Mixture Supervised Fine-Tuning Eliciting Efficient Reasoning in Large Language Modelscited by 37
  32. 2025ThinkSwitcher: When to Think Hard, When to Think Fastcited by 37
  33. 2025Optimizing Length Compression in Large Reasoning Modelscited by 37
  34. 2025Don't Overthink It: A Survey of Efficient R1-style Large Reasoning Modelscited by 36
  35. 2025Z1: Efficient Test-time Scaling with Codecited by 35
  36. 2025DLER: Doing Length pEnalty Right - Incentivizing More Intelligence per Token via Reinforcement Learningcited by 34
  37. 2025ConCISE: Confidence-guided Compression in Step-by-step Efficient Reasoningcited by 33
  38. 2025Scalable Chain of Thoughts via Elastic Reasoningcited by 32

50 most cited of 278 All 278 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025Think Before You Prune: Self-Reflective Structured Pruning for Reasoning Language Models
  2. 2024A Comparative Study on Reasoning Patterns of OpenAI's o1 Model
  3. 2025Thoughts Are All Over the Place: On the Underthinking of o1-Like LLMs
  4. 2025Think, Prune, Train, Improve: Scaling Reasoning without Scaling Models
  5. 2025Double-Checker: Enhancing Reasoning of Slow-Thinking LLMs via Self-Critical Fine-Tuning
  6. 2025Towards Concise and Adaptive Thinking in Large Reasoning Models: A Survey
  7. 2025Think Clearly: Improving Reasoning via Redundant Token Pruning
  8. 2025Inference-Time Computations for LLM Reasoning and Planning: A Benchmark and Insights