Pith. sign in

REVIEW 6 cited by

Optimizing Length Compression in Large Reasoning Models

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2506.14755 v2 pith:B7XZR6ZN submitted 2025-06-17 cs.AI cs.CL

Zhengxiang Cheng , Dongping Chen , Mingyang Fu , Tianyi Zhou This is my paper
classification cs.AIcs.CL
keywords lc-r1reasoninglengthmodelsprinciplescompressioninvalidlarge
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Large Reasoning Models (LRMs) have achieved remarkable success, yet they often suffer from producing unnecessary and verbose reasoning chains. We identify a core aspect of this issue as "invalid thinking" -- models tend to repeatedly double-check their work after having derived the correct answer. To address this specific inefficiency, we move beyond the general principles of Efficacy and Efficiency to propose two new, fine-grained principles: Brevity, which advocates for eliminating redundancy, and Sufficiency, which ensures critical reasoning steps are preserved. Guided by these principles, we introduce LC-R1, a post-training method based on Group Relative Policy Optimization (GRPO). LC-R1 employs a novel combination of a Length Reward for overall conciseness and a Compress Reward that is specifically designed to remove the invalid portion of the thinking process. Extensive experiments on multiple reasoning benchmarks demonstrate that LC-R1 achieves a significant reduction in sequence length (~50%) with only a marginal (~2%) drop in accuracy, achieving a favorable trade-off point on the Pareto frontier that prioritizes high compression. Our analysis further validates the robustness of LC-R1 and provides valuable insights for developing more powerful yet computationally efficient LRMs. Our code is released at https://github.com/zxiangx/LC-R1.

Discussion (0). Sign in to comment.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. EvoThink: Evolving Thinking in Large Reasoning Models via Self-Pruning and Aha-Moment Preference Optimization

    cs.AI 2026-07 conditional novelty 6.0 of 10

    EvoThink reduces LRM overthinking by pruning redundant atomic reasoning steps (SPT) and training on diversity-selected wrong-to-right mutation data (AMPO), cutting tokens and improving math/code accuracy.

  2. Compress the Easy, Explore the Hard: Difficulty-Aware Entropy Regularization for Efficient LLM Reasoning

    cs.LG 2026-02 conditional novelty 6.0 of 10

    CEEH selectively applies entropy regularization to hard questions and a shortest-correct-length penalty to easy ones, reducing reasoning length while preserving accuracy across six math benchmarks.

  3. ConPress: Learning Efficient Reasoning from Multi-Question Contextual Pressure

    cs.CL 2026-02 conditional novelty 6.0 of 10

    Multi-question prompts elicit shorter chain-of-thought traces, and fine-tuning on those traces transfers the compression to single-question reasoning.

  4. Reconsidering Overthinking: Penalizing Internal and External Redundancy in CoT Reasoning

    cs.AI 2025-08 conditional novelty 6.0 of 10

    A dual-penalty RL method that compresses chain-of-thought traces by separately penalizing internal semantic stagnation and external post-answer continuation reduces reasoning length by about 40% while preserving accur...

  5. Failure Cases Are Better Learned But Boundary Says Sorry: Facilitating Smooth Perception Change for Accuracy-Robustness Trade-Off in Adversarial Training

    cs.CV 2025-08 reject novelty 5.0 of 10

    A paper whose abstract describes new adversarial training experiments, but whose full text is a different paper on CoT compression, leaving the claims unsupported.

  6. Towards Concise and Adaptive Thinking in Large Reasoning Models: A Survey

    cs.AI 2025-07 conditional novelty 3.0 of 10

    A comprehensive review that categorizes methods for shortening and adaptively triggering chain-of-thought reasoning in large language models.

Reference graph

Works this paper leans on

42 extracted references · 5 canonical work pages · cited by 6 Pith papers

  1. [1]

    Openai o1 system card

    Aaron Jaech, Adam Kalai, Adam Lerer, Adam Richardson, Ahmed El-Kishky, Aiden Low, Alec Helyar, Aleksander Madry, Alex Beutel, Alex Carney, et al. Openai o1 system card. arXiv preprint arXiv:2412.16720, 2024

    arXiv 2024

  2. [2]

    DeepSeek-AI, Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Ruoyu Zhang, Runxin Xu, Qihao Zhu, Shirong Ma, Peiyi Wang, Xiao Bi, Xiaokang Zhang, Xingkai Yu, Yu Wu, Z. F. Wu, Zhibin Gou, Zhihong Shao, Zhuoshu Li, Ziyi Gao ..., and Zhen Zhang. Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning, 2025. URL https://arxiv.org/...

    arXiv 2025

  3. [3]

    Chain-of-thought prompting elicits reasoning in large language models, 2023

    Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Brian Ichter, Fei Xia, Ed Chi, Quoc Le, and Denny Zhou. Chain-of-thought prompting elicits reasoning in large language models, 2023. URL https://arxiv.org/abs/2201.11903

    arXiv 2023

  4. [4]

    Challenging the boundaries of reasoning: An olympiad-level math benchmark for large language models, 2025

    Haoxiang Sun, Yingqian Min, Zhipeng Chen, Wayne Xin Zhao, Zheng Liu, Zhongyuan Wang, Lei Fang, and Ji-Rong Wen. Challenging the boundaries of reasoning: An olympiad-level math benchmark for large language models, 2025. URL https://arxiv.org/abs/2503.21380

    arXiv 2025

  5. [5]

    Alex Gu, Baptiste Rozière, Hugh Leather, Armando Solar-Lezama, Gabriel Synnaeve, and Sida I. Wang. Cruxeval: A benchmark for code reasoning, understanding and execution. arXiv preprint arXiv:2401.03065, 2024

    arXiv 2024

  6. [6]

    Towards reasoning era: A survey of long chain-of-thought for reasoning large language models, 2025

    Qiguang Chen, Libo Qin, Jinhao Liu, Dengyun Peng, Jiannan Guan, Peng Wang, Mengkang Hu, Yuhang Zhou, Te Gao, and Wanxiang Che. Towards reasoning era: A survey of long chain-of-thought for reasoning large language models, 2025. URL https://arxiv.org/abs/2503.09567

    arXiv 2025

  7. [7]

    L1: Controlling how long a reasoning model thinks with reinforcement learning, 2025

    Pranjal Aggarwal and Sean Welleck. L1: Controlling how long a reasoning model thinks with reinforcement learning, 2025. URL https://arxiv.org/abs/2503.04697

    arXiv 2025

  8. [8]

    Do not think that much for 2+ 3=? on the overthinking of o1-like llms

    Xingyu Chen, Jiahao Xu, Tian Liang, Zhiwei He, Jianhui Pang, Dian Yu, Linfeng Song, Qiuzhi Liu, Mengfei Zhou, Zhuosheng Zhang, et al. Do not think that much for 2+ 3=? on the overthinking of o1-like llms. arXiv preprint arXiv:2412.21187, 2024

    arXiv 2024

Show all 42 references
  1. [9]

    Stop overthinking: A survey on efficient reasoning for large language models

    Yang Sui, Yu-Neng Chuang, Guanchu Wang, Jiamu Zhang, Tianyi Zhang, Jiayi Yuan, Hongyi Liu, Andrew Wen, Hanjie Chen, Xia Hu, et al. Stop overthinking: A survey on efficient reasoning for large language models. arXiv preprint arXiv:2503.16419, 2025

    2025 arXiv

  2. [10]

    The danger of overthinking: Examining the reasoning-action dilemma in agentic tasks

    Alejandro Cuadron, Dacheng Li, Wenjie Ma, Xingyao Wang, Yichuan Wang, Siyuan Zhuang, Shu Liu, Luis Gaspar Schroeder, Tian Xia, Huanzhi Mao, et al. The danger of overthinking: Examining the reasoning-action dilemma in agentic tasks. arXiv preprint arXiv:2502.08235, 2025

    2025 arXiv

  3. [11]

    Adar1: From long-cot to hybrid-cot via bi-level adaptive reasoning optimization

    Haotian Luo, Haiying He, Yibo Wang, Jinluan Yang, Rui Liu, Naiqiang Tan, Xiaochun Cao, Dacheng Tao, and Li Shen. Adar1: From long-cot to hybrid-cot via bi-level adaptive reasoning optimization. arXiv preprint arXiv:2504.21659, 2025 a

    2025 arXiv

  4. [12]

    Dast: Difficulty-adaptive slow-thinking for large reasoning models, 2025

    Yi Shen, Jian Zhang, Jieyun Huang, Shuming Shi, Wenjing Zhang, Jiangze Yan, Ning Wang, Kai Wang, and Shiguo Lian. Dast: Difficulty-adaptive slow-thinking for large reasoning models, 2025. URL https://arxiv.org/abs/2503.04472

    2025

  5. [13]

    Thinkprune: Pruning long chain-of-thought of llms via reinforcement learning

    Bairu Hou, Yang Zhang, Jiabao Ji, Yujian Liu, Kaizhi Qian, Jacob Andreas, and Shiyu Chang. Thinkprune: Pruning long chain-of-thought of llms via reinforcement learning. arXiv preprint arXiv:2504.01296, 2025

    2025 arXiv

  6. [14]

    O1-pruner: Length-harmonizing fine-tuning for o1-like reasoning pruning, 2025 b

    Haotian Luo, Li Shen, Haiying He, Yibo Wang, Shiwei Liu, Wei Li, Naiqiang Tan, Xiaochun Cao, and Dacheng Tao. O1-pruner: Length-harmonizing fine-tuning for o1-like reasoning pruning, 2025 b . URL https://arxiv.org/abs/2501.12570

    2025 arXiv

  7. [15]

    Kimi Team, Angang Du, Bofei Gao, Bowei Xing, Changjiu Jiang, Cheng Chen, Cheng Li, Chenjun Xiao, Chenzhuang Du, Chonghua Liao, Chuning Tang, Congcong Wang, Dehao Zhang, Enming Yuan, Enzhe Lu, Fengxiang Tang, Flood Sung, Guangda Wei, Guokun Lai, Haiqing Guo, Han Zhu, Hao Ding, ...

    2025 arXiv

  8. [16]

    Qwen3, April 2025 a

    Qwen Team. Qwen3, April 2025 a . URL https://qwenlm.github.io/blog/qwen3/

    2025

  9. [17]

    Qwq-32b: Embracing the power of reinforcement learning, March 2025 b

    Qwen Team. Qwq-32b: Embracing the power of reinforcement learning, March 2025 b . URL https://qwenlm.github.io/blog/qwq-32b/

    2025

  10. [18]

    Yian Zhang, and Chris Alexiuk

    Akhiad Bercovich, Itay Levy, Izik Golan, Mohammad Dabbah, Ran El-Yaniv, Omri Puny, Ido Galil, Zach Moshe, Tomer Ronen, Najeeb Nabwani, Ido Shahaf, Oren Tropp, Ehud Karpas, Ran Zilberstein, Jiaqi Zeng, Soumye Singhal, Alexander Bukharin, ... Yian Zhang, and Chris Alexiuk. Llama...

    2025

  11. [19]

    Understanding r1-zero-like training: A critical perspective, 2025

    Zichen Liu, Changyu Chen, Wenjun Li, Penghui Qi, Tianyu Pang, Chao Du, Wee Sun Lee, and Min Lin. Understanding r1-zero-like training: A critical perspective, 2025. URL https://arxiv.org/abs/2503.20783

    2025 arXiv

  12. [20]

    Dapo: An open-source llm reinforcement learning system at scale

    Qiying Yu, Zheng Zhang, Ruofei Zhu, Yufeng Yuan, Xiaochen Zuo, Yu Yue, Tiantian Fan, Gaohong Liu, Lingjun Liu, Xin Liu, et al. Dapo: An open-source llm reinforcement learning system at scale. arXiv preprint arXiv:2503.14476, 2025

    2025 arXiv

  13. [21]

    Direct preference optimization: Your language model is secretly a reward model

    Rafael Rafailov, Archit Sharma, Eric Mitchell, Christopher D Manning, Stefano Ermon, and Chelsea Finn. Direct preference optimization: Your language model is secretly a reward model. Advances in Neural Information Processing Systems, 36: 0 53728--53741, 2023

    2023

  14. [22]

    OpenAI. Chatgpt. https://openai.com/o1/, 2024

    2024

  15. [23]

    Gemini 2.5 pro

    Google. Gemini 2.5 pro. https://cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/2-5-pro, 2025 a

    2025

  16. [24]

    Phi-4-reasoning technical report, 2025

    Marah Abdin, Sahaj Agarwal, Ahmed Awadallah, Vidhisha Balachandran, Harkirat Behl, Lingjiao Chen, Gustavo de Rosa, Suriya Gunasekar, Mojan Javaheripi, Neel Joshi, Piero Kauffmann, Yash Lara, Caio César Teodoro Mendes, Arindam Mitra, Besmira Nushi, Dimitris Papailiopoulos, Olli...

    2025 arXiv

  17. [25]

    Zhihong Shao, Peiyi Wang, Qihao Zhu, Runxin Xu, Junxiao Song, Xiao Bi, Haowei Zhang, Mingchuan Zhang, Y. K. Li, Y. Wu, and Daya Guo. Deepseekmath: Pushing the limits of mathematical reasoning in open language models, 2024. URL https://arxiv.org/abs/2402.03300

    2024 arXiv

  18. [26]

    Code-r1: Reproducing r1 for code with reliable rewards

    Jiawei Liu and Lingming Zhang. Code-r1: Reproducing r1 for code with reliable rewards. 2025

    2025

  19. [27]

    Reinforce++: An efficient rlhf algorithm with robustness to both prompt and reward models, 2025

    Jian Hu, Jason Klein Liu, and Wei Shen. Reinforce++: An efficient rlhf algorithm with robustness to both prompt and reward models, 2025. URL https://arxiv.org/abs/2501.03262

    2025 arXiv

  20. [28]

    Cot-valve: Length-compressible chain-of-thought tuning, 2025 a

    Xinyin Ma, Guangnian Wan, Runpeng Yu, Gongfan Fang, and Xinchao Wang. Cot-valve: Length-compressible chain-of-thought tuning, 2025 a . URL https://arxiv.org/abs/2502.09601

    2025 arXiv

  21. [29]

    Training language models to reason efficiently, 2025

    Daman Arora and Andrea Zanette. Training language models to reason efficiently, 2025. URL https://arxiv.org/abs/2502.04463

    2025

  22. [30]

    Aytes, Jinheon Baek, and Sung Ju Hwang

    Simon A. Aytes, Jinheon Baek, and Sung Ju Hwang. Sketch-of-thought: Efficient llm reasoning with adaptive cognitive-inspired sketching, 2025. URL https://arxiv.org/abs/2503.05179

    2025

  23. [31]

    Token-budget-aware llm reasoning

    Tingxu Han, Zhenting Wang, Chunrong Fang, Shiyu Zhao, Shiqing Ma, and Zhenyu Chen. Token-budget-aware llm reasoning. arXiv preprint arXiv:2412.18547, 2024

    2024 arXiv

  24. [32]

    Reasoning models can be effective without thinking, 2025 b

    Wenjie Ma, Jingxuan He, Charlie Snell, Tyler Griggs, Sewon Min, and Matei Zaharia. Reasoning models can be effective without thinking, 2025 b . URL https://arxiv.org/abs/2504.09858

    2025 arXiv

  25. [33]

    Qwen2.5: A party of foundation models, September 2024

    Qwen Team. Qwen2.5: A party of foundation models, September 2024. URL https://qwenlm.github.io/blog/qwen2.5/

    2024

  26. [34]

    Gemini 2.5 flash

    Google. Gemini 2.5 flash. https://developers.googleblog.com/en/start-building-with-gemini-25-flash/, 2025 b

    2025

  27. [35]

    Distilling system 2 into system 1, 2024

    Ping Yu, Jing Xu, Jason Weston, and Ilia Kulikov. Distilling system 2 into system 1, 2024. URL https://arxiv.org/abs/2407.06023

    2024 arXiv

  28. [36]

    The 23rd international conference on artificial intelligence in medicine (aime 2025)

    International Conference on Artificial Intelligence in Medicine . The 23rd international conference on artificial intelligence in medicine (aime 2025). https://aime25.aimedicine.info/. Accessed: 2025-06-10

    2025

  29. [37]

    Let's verify step by step

    Hunter Lightman, Vineet Kosaraju, Yuri Burda, Harrison Edwards, Bowen Baker, Teddy Lee, Jan Leike, John Schulman, Ilya Sutskever, and Karl Cobbe. Let's verify step by step. In The Twelfth International Conference on Learning Representations, 2023

    2023

  30. [38]

    Training verifiers to solve math word problems

    Karl Cobbe, Vineet Kosaraju, Mohammad Bavarian, Mark Chen, Heewoo Jun, Lukasz Kaiser, Matthias Plappert, Jerry Tworek, Jacob Hilton, Reiichiro Nakano, Christopher Hesse, and John Schulman. Training verifiers to solve math word problems. arXiv preprint arXiv:2110.14168, 2021

    arXiv

  31. [39]

    American mathematics competitions (amc)

    Mathematical Association of America . American mathematics competitions (amc). https://maa-amc.org/student-programs/amc/. Accessed: 2025-06-10

    2025

  32. [40]

    Gpqa: A graduate-level google-proof q&a benchmark

    David Rein, Betty Li Hou, Asa Cooper Stickland, Jackson Petty, Richard Yuanzhe Pang, Julien Dirani, Julian Michael, and Samuel R Bowman. Gpqa: A graduate-level google-proof q&a benchmark. In First Conference on Language Modeling, 2024

    2024

  33. [41]

    Livecodebench: Holistic and contamination free evaluation of large language models for code

    Naman Jain, King Han, Alex Gu, Wen-Ding Li, Fanjia Yan, Tianjun Zhang, Sida Wang, Armando Solar-Lezama, Koushik Sen, and Ion Stoica. Livecodebench: Holistic and contamination free evaluation of large language models for code. arXiv preprint arXiv:2403.07974, 2024. doi:10.48550...

    2024 doi arXiv

  34. [42]

    Trl: Transformer reinforcement learning

    Leandro von Werra, Younes Belkada, Lewis Tunstall, Edward Beeching, Tristan Thrush, Nathan Lambert, Shengyi Huang, Kashif Rasul, and Quentin Gallouédec. Trl: Transformer reinforcement learning. https://github.com/huggingface/trl, 2020

    2020

Pith tools