Elamur Research⌕
☰
home / papers / 2025 / Efficient Reasoning Models: A Survey

Efficient Reasoning Models: A Survey

Paper · 2025 · Sicheng Feng, Gongfan Fang, Xinyin Ma et al.

Summary

Large Reasoning Models such as OpenAI o1 and DeepSeek R1 solve hard math and coding problems by writing long Chain-of-Thoughts (CoTs) before answering. This "slow thinking" is costly. Models often overthink trivial questions, longer chains can lower accuracy, and very large models need heavy compute and memory. The authors review recent work on making reasoning cheaper. They organize it by the kind of redundancy each method removes:

  • Shorter: compressing CoTs via RL with length penalties, SFT on variable-length CoT data, prompt-driven control or routing, and latent reasoning.
  • Smaller: building compact models through distillation, quantization and pruning, or RL applied directly to small models.
  • Faster: cheaper decoding, such as efficient sampling, efficient self-consistency, and problem decomposition.

The survey also covers metrics and benchmarks, open challenges, and how its goal-driven taxonomy differs from surveys organized by training stage or by technique. It gives a structured map of the field for anyone deploying reasoning models under tight compute budgets.

Key points

  • The taxonomy has three branches. Shorter includes L1, O1-Pruner, TokenSkip, CoT-Valve and Coconut. Smaller includes CoT-KD, Open-RS and DeepScaleR. Faster includes Fast Best-of-N, ESC and SoT. Paper lists are maintained in the GitHub repo Awesome-Efficient-Reasoning-Models.
  • In an illustrative example, a model used 619 tokens to answer what 1 plus 2 equals, which shows the overthinking problem.
  • Model size is a major cost driver: DeepSeek R1 has 685B parameters. In the compression studies reviewed, quantization largely preserved reasoning, while early pruning results were weak.
  • The authors measured token count against latency for Qwen2.5-7B on MATH-500 (written "MAHT-500" in the extracted text). The Pearson correlation was 0.9998, which supports token count as a proxy for runtime.
  • Data efficiency matters. For example, s1 trains on a curated set of just 1,000 examples selected for difficulty, diversity and quality.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓National University of SingaporeSingapore · Singapore180 works · 34,304 citations🎓Nankai UniversityTianjin · China24 works · 762 citations

 Fields, topics and concepts

📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 worksComputer Vision and Pattern Recognition · 2,869 works

🧭 Topics 1Research subjects and languages the work is about.Reasoning · 1,426 works

Cite

Cite as Sicheng Feng, Gongfan Fang, Xinyin Ma and Xinchao Wang. 2025. Efficient Reasoning Models: A Survey. arXiv:2504.10903.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.48550/arxiv.2504.10903 · arXiv 2504.10903

@misc{feng2025efficient,
  title = {{Efficient Reasoning Models: A Survey}},
  author = {Sicheng Feng, Gongfan Fang, Xinyin Ma and Xinchao Wang},
  year = {2025},
  doi = {10.48550/arxiv.2504.10903},
  eprint = {2504.10903},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2504.10903}
}

References

199 references, 61 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Pranjal Aggarwal and Sean Welleck. L1: Controlling how long a reasoning model thinks with reinforcement learning. arXiv preprint arXiv:2503.04697, 2025.
  2. Pranjal Aggarwal, Aman Madaan, Yiming Yang, et al. Let’s sample step by step: Adaptive-consistency for efficient reasoning and coding with llms. arXiv preprint arXiv:2305.11860, 2023. find
  3. Open AI. Introducing openai o1-preview. 2024. find
  4. Lasse F Wolff Anthony, Benjamin Kanding, and Raghavendra Selvan. Carbontracker: Tracking and predicting the carbon footprint of training deep learning models. arXiv preprint arXiv:2007.03051, 2020. — [arXiv:2007.03051](https://arxiv.org/abs/2007.03051) find
  5. Anthropic. Claude 3.7 sonnet. 2025. find
  6. Daman Arora and Andrea Zanette. arXiv:2502.04463, 2025. find
  7. Training language models to reason efficiently. arXiv preprint Berk Atil, Alexa Chittams, Liseng Fu, Ferhan Ture, Lixinyu Xu, and Breck Baldwin. Llm stability: A detailed analysis with some surprises. arXiv preprint arXiv:2408.04667, 2024. find
  8. Simon A Aytes, Jinheon Baek, and Sung Ju Hwang. Sketch-of-thought: Efficient llm reasoning with adaptive cognitive-inspired sketching. arXiv preprint arXiv:2503.05179, 2025.
  9. Maciej Besta, Nils Blach, Ales Kubicek, Robert Gerstenberger, Michal Podstawski, Lukas Gianinazzi, Joanna Gajda, Tomasz Lehmann, Hubert Niewiadomski, Piotr Nyczyk, et al. Graph of thoughts: Solving elaborate problems with large language models. In AAAI, 2024. find
  10. Mark Chen, Jerry Tworek, Heewoo Jun, Qiming Yuan, Henrique Ponde De Oliveira Pinto, Jared Kaplan, Harri Edwards, Yuri Burda, Nicholas Joseph, Greg Brockman, et al. Evaluating large language models trained on code. arXiv preprint arXiv:2107.03374, 2021.
  11. Qiguang Chen, Libo Qin, Jiaqi Wang, Jingxuan Zhou, and Wanxiang Che. Unlocking the capabilities of thought: A reasoning boundary framework to quantify and optimize chain-of-thought. In NeurIPS, 2024a. find
  12. 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. arXiv preprint arXiv:2503.09567, 2025a.
  13. Wenhu Chen, Xueguang Ma, Xinyi Wang, and William W Cohen. Program of thoughts prompting: Disentangling computation from reasoning for numerical reasoning tasks. arXiv preprint arXiv:2211.12588, 2022. find
  14. Xiaoshu Chen, Sihang Zhou, Ke Liang, and Xinwang Liu. Distilling reasoning ability from large language models with adaptive thinking. arXiv preprint arXiv:2404.09170, 2024b. — [arXiv:2404.09170](https://arxiv.org/abs/2404.09170) find
  15. Xinghao Chen, Zhijing Sun, Wenjin Guo, Miaoran Zhang, Yanjun Chen, Yirong Sun, Hui Su, Yijie Pan, Dietrich Klakow, Wenjie Li, et al. Unveiling the key factors for distilling chain-of-thought reasoning. arXiv preprint arXiv:2502.18001, 2025b. find
  16. 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, 2024c.
  17. Xinyun Chen, Maxwell Lin, Nathanael Schärli, and Denny Zhou. Teaching large language models to self-debug. In ICLR, 2024d. find
  18. Jeffrey Cheng and Benjamin Van Durme. Compressed chain of thought: Efficient reasoning through dense representations. arXiv preprint arXiv:2412.13171, 2024. find
  19. Yu-Neng Chuang, Helen Zhou, Prathusha Sarma, Parikshit Gopalan, John Boccio, Sara Bolouki, and Xia Hu. Learning to route llms with confidence tokens. arXiv preprint arXiv:2410.13284, 2024. — [arXiv:2410.13284](https://arxiv.org/abs/2410.13284) find
  20. Yu-Neng Chuang, Leisheng Yu, Guanchu Wang, Lizhe Zhang, Zirui Liu, Xuanting Cai, Yang Sui, Vladimir Braverman, and Xia Hu. Confident or seek stronger: Exploring uncertainty-based on-device llm routing from benchmarking to generalization. arXiv preprint arXiv:2502.04428, 2025. — [arXiv:2502.04428](https://arxiv.org/abs/2502.04428) find
  21. Karl Cobbe, Vineet Kosaraju, Mohammad Bavarian, Mark Chen, Heewoo Jun, Lukasz Kaiser, Matthias Plappert, Jerry Tworek, Jacob Hilton, Reiichiro Nakano, et al. Training verifiers to solve math word problems. arXiv preprint arXiv:2110.14168, 2021.
  22. Rémi Coulom. Efficient selectivity and backup operators in monte-carlo tree search. In International conference on computers and games, 2006. find
  23. 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. find
  24. Can Cui, Yunsheng Ma, Xu Cao, Wenqian Ye, Yang Zhou, Kaizhao Liang, Jintai Chen, Juanwu Lu, Zichong Yang, Kuei-Da Liao, et al. A survey on multimodal large language models for autonomous driving. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 2024. find
  25. Yingqian Cui, Pengfei He, Jingying Zeng, Hui Liu, Xianfeng Tang, Zhenwei Dai, Yan Han, Chen Luo, Jing Huang, Zhen Li, et al. Stepwise perplexity-guided refinement for efficient chain-of-thought reasoning in large language models. arXiv preprint arXiv:2502.13260, 2025. find
  26. Quy-Anh Dang and Chris Ngo. Reinforcement learning for reasoning in small llms: What works and what doesn’t. arXiv preprint arXiv:2503.16219, 2025. find
  27. Yuntian Deng, Kiran Prasad, Roland Fernandez, Paul Smolensky, Vishrav Chaudhary, and Stuart Shieber. Implicit chain of thought reasoning via knowledge distillation. arXiv preprint arXiv:2311.01460, 2023. find
  28. Yuntian Deng, Yejin Choi, and Stuart Shieber. From explicit cot to implicit cot: Learning to internalize cot step by step. arXiv preprint arXiv:2405.14838, 2024. find
  29. Mengru Ding, Hanmeng Liu, Zhizhang Fu, Jian Song, Wenbo Xie, and Yue Zhang. Break the chain: Large language models can be shortcut reasoners. arXiv preprint arXiv:2406.06580, 2024. — [arXiv:2406.06580](https://arxiv.org/abs/2406.06580) find
  30. Ruomeng Ding, Chaoyun Zhang, Lu Wang, Yong Xu, Minghua Ma, Wei Zhang, Si Qin, Saravan Rajmohan, Qingwei Lin, and Dongmei Zhang. Everything of thoughts: Defying the law of penrose triangle for thought generation. arXiv preprint arXiv:2311.04254, 2023. find
  31. Yifu Ding, Wentao Jiang, Shunyu Liu, Yongcheng Jing, Jinyang Guo, Yingjie Wang, Jing Zhang, Zengmao Wang, Ziwei Liu, Bo Du, et al. Dynamic parallel tree search for efficient llm reasoning. arXiv preprint arXiv:2502.16235, 2025.
  32. Jiafei Duan, Samson Yu, Hui Li Tan, Hongyuan Zhu, and Cheston Tan. A survey of embodied ai: From simulators to research tasks. IEEE Transactions on Emerging Topics in Computational Intelligence, 6(2): 230–244, 2022. find
  33. Gongfan Fang, Xinyin Ma, Mingli Song, Michael Bi Mi, and Xinchao Wang. Depgraph: Towards any structural pruning. In CVPR, 2023. find
  34. Gongfan Fang, Xinyin Ma, Michael Bi Mi, and Xinchao Wang. Isomorphic pruning for vision models. In ECCV, 2024. find
  35. Gongfan Fang, Xinyin Ma, and Xinchao Wang. Thinkless: Llm learns when to think. arXiv preprint arXiv:2505.13379, 2025. — [arXiv:2505.13379](https://arxiv.org/abs/2505.13379) find
  36. Sicheng Feng, Siyu Li, Luonan Chen, and Shengquan Chen. Unveiling potential threats: backdoor attacks in single-cell pre-trained models. Cell Discovery, 10(1):122, 2024a. find
  37. Sicheng Feng, Keda Tao, and Huan Wang. arXiv:2412.00143, 2024b. — [arXiv:2412.00143](https://arxiv.org/abs/2412.00143) find
  38. Is oracle pruning the true oracle? arXiv preprint Sicheng Feng, Song Wang, Shuyi Ouyang, Lingdong Kong, Zikai Song, Jianke Zhu, Huan Wang, and Xinchao Wang. Can mllms guide me home? a benchmark study on fine-grained visual reasoning from transit maps. arXiv preprint arXiv:2505.18675, 2025. — [arXiv:2505.18675](https://arxiv.org/abs/2505.18675) find
  39. Tao Feng, Yicheng Li, Li Chenglin, Hao Chen, Fei Yu, and Yin Zhang. Teaching small language models reasoning through counterfactual distillation. In EMNLP, 2024c.
  40. Elias Frantar, Saleh Ashkboos, Torsten Hoefler, and Dan Alistarh. GPTQ: Accurate post-training compression for generative pretrained transformers. In ICLR, 2023a. find
  41. Elias Frantar, Saleh Ashkboos, Torsten Hoefler, and Dan Alistarh. Gptq: Accurate post-training quantization for generative pre-trained transformers. In ICLR, 2023b. find
  42. Peizhong Gao, Ao Xie, Shaoguang Mao, Wenshan Wu, Yan Xia, Haipeng Mi, and Furu Wei. Meta reasoning for large language models. arXiv preprint arXiv:2406.11698, 2024. — [arXiv:2406.11698](https://arxiv.org/abs/2406.11698) find
  43. Mor Geva, Daniel Khashabi, Elad Segal, Tushar Khot, Dan Roth, and Jonathan Berant. Did aristotle use a laptop? a question answering benchmark with implicit reasoning strategies. Transactions of the Association for Computational Linguistics, 2021. find
  44. Angeliki Giannou, Shashank Rajput, Jy-yong Sohn, Kangwook Lee, Jason D Lee, and Dimitris Papailiopoulos. Looped transformers as programmable computers. In ICML, 2023. find
  45. Vinod Goel. Sketches of thought. MIT press, 1995. find
  46. Zhibin Gou, Zhihong Shao, Yeyun Gong, Yelong Shen, Yujiu Yang, Nan Duan, and Weizhu Chen. Critic: Large language models can self-correct with tool-interactive critiquing. In ICLR, 2024.
  47. Robert M. Gray and David L. Neuhoff. Quantization. IEEE transactions on information theory, 1998. find
  48. Albert Gu and Tri Dao. Mamba: Linear-time sequence modeling with selective state spaces. In COLM, 2024. find
  49. Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Ruoyu Zhang, Runxin Xu, Qihao Zhu, Shirong Ma, Peiyi Wang, Xiao Bi, et al. Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning. arXiv preprint arXiv:2501.12948, 2025.
  50. Song Han, Huizi Mao, and William J Dally. Deep compression: Compressing deep neural networks with pruning, trained quantization and huffman coding. In ICLR, 2016. find
  51. Tingxu Han, Zhenting Wang, Chunrong Fang, Shiyu Zhao, Shiqing Ma, and Zhenyu Chen. Token-budget-aware llm reasoning. arXiv preprint arXiv:2412.18547, 2024.
  52. Jitai Hao, Yuke Zhu, Tian Wang, Jun Yu, Xin Xin, Bo Zheng, Zhaochun Ren, and Sheng Guo. Omnikv: Dynamic context selection for efficient long-context llms. In ICLR, 2025. find
  53. Shibo Hao, Sainbayar Sukhbaatar, DiJia Su, Xian Li, Zhiting Hu, Jason Weston, and Yuandong Tian. Training large language models to reason in a continuous latent space. arXiv preprint arXiv:2412.06769, 2024. find
  54. Masoud Hashemi, Oluwanifemi Bamgbose, Sathwik Tejaswi Madhusudhan, Jishnu Sethumadhavan Nair, Aman Tiwari, and Vikas Yadav. Dna bench: When silence is smarter–benchmarking over-reasoning in reasoning llms. arXiv preprint arXiv:2503.15793, 2025. — [arXiv:2503.15793](https://arxiv.org/abs/2503.15793) find
  55. Dan Hendrycks, Collin Burns, Saurav Kadavath, Akul Arora, Steven Basart, Eric Tang, Dawn Song, and Jacob Steinhardt. Measuring mathematical problem solving with the math dataset. arXiv preprint arXiv:2103.03874, 2021.
  56. Geoffrey Hinton, Oriol Vinyals, and Jeff Dean. Distilling the knowledge in a neural network. arXiv preprint arXiv:1503.02531, 2015.
  57. 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.
  58. Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, Weizhu Chen, et al. Lora: Low-rank adaptation of large language models. In ICLR, 2022.
  59. Hanxu Hu, Hongyuan Lu, Huajian Zhang, Yun-Ze Song, Wai Lam, and Yue Zhang. Chain-of-symbol prompting for spatial reasoning in large language models. In First Conference on Language Modeling, 2024. find
  60. Yangliu Hu, Zikai Song, Na Feng, Yawei Luo, Junqing Yu, Yi-Ping Phoebe Chen, and Wei Yang. Sf2t: Self-supervised fragment finetuning of video-llms for fine-grained understanding. arXiv preprint arXiv:2504.07745, 2025. — [arXiv:2504.07745](https://arxiv.org/abs/2504.07745) find
  61. Chengsong Huang, Langlin Huang, Jixuan Leng, Jiacheng Liu, and Jiaxin Huang. Efficient test-time scaling via self-calibration. arXiv preprint arXiv:2503.00031, 2025. — [arXiv:2503.00031](https://arxiv.org/abs/2503.00031) find
  62. 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.
  63. Mingyu Jin, Weidi Luo, Sitao Cheng, Xinyi Wang, Wenyue Hua, Ruixiang Tang, William Yang Wang, and Yongfeng Zhang. Disentangling memory and reasoning ability in large language models. arXiv preprint arXiv:2411.13504, 2024a.
  64. Mingyu Jin, Qinkai Yu, Dong Shu, Haiyan Zhao, Wenyue Hua, Yanda Meng, Yongfeng Zhang, and Mengnan Du. The impact of reasoning step length on large language models. arXiv preprint arXiv:2401.04925, 2024b.
  65. Shuowei Jin, Yongji Wu, Haizhong Zheng, Qingzhao Zhang, Matthew Lentz, Z Morley Mao, Atul Prakash, Feng Qian, and Danyang Zhuo. Adaptive skeleton graph decoding. arXiv preprint arXiv:2402.12280, 2024c. — [arXiv:2402.12280](https://arxiv.org/abs/2402.12280) find
  66. Yu Kang, Xianghui Sun, Liangyu Chen, and Wei Zou. C3ot: Generating shorter chain-of-thought without compromising effectiveness. arXiv preprint arXiv:2412.11664, 2024. find
  67. Aybora Koksal and Aydin Alatan Alatan. Milchat: Introducing chain of thought reasoning and grpo to a multimodal small language model for remote sensing. arXiv preprint arXiv:2505.07984, 2025. — [arXiv:2505.07984](https://arxiv.org/abs/2505.07984) find
  68. Martin Kuo, Jianyi Zhang, Aolin Ding, Qinsi Wang, Louis DiValentin, Yujia Bao, Wei Wei, Da-Cheng Juan, Hai Li, and Yiran Chen. H-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 thinking. arX… — [arXiv:2502.12893](https://arxiv.org/abs/2502.12893) find
  69. Alexandre Lacoste, Alexandra Luccioni, Victor Schmidt, and Thomas Dandres. Quantifying the carbon emissions of machine learning. arXiv preprint arXiv:1910.09700, 2019.
  70. Yann LeCun, John Denker, and Sara Solla. Optimal brain damage. In NeurIPS, 1989. find
  71. Ayeong Lee, Ethan Che, and Tianyi Peng. How well do llms compress their own chain-of-thought? a token complexity approach. arXiv preprint arXiv:2503.01141, 2025. — [arXiv:2503.01141](https://arxiv.org/abs/2503.01141) find
  72. Chen Li, Nazhou Liu, and Kai Yang. Adaptive group policy optimization: Towards stable training and token-efficient reasoning. arXiv preprint arXiv:2503.15952, 2025a. — [arXiv:2503.15952](https://arxiv.org/abs/2503.15952) find
  73. Chenglin Li, Qianglong Chen, Liangyue Li, Caiyu Wang, Yicheng Li, Zulong Chen, and Yin Zhang. Mixed distillation helps smaller language model better reasoning. arXiv preprint arXiv:2312.10730, 2023a.
  74. Peiji Li, Kai Lv, Yunfan Shao, Yichuan Ma, Linyang Li, Xiaoqing Zheng, Xipeng Qiu, and Qipeng Guo. Fastmcts: A simple sampling strategy for data synthesis. arXiv preprint arXiv:2502.11476, 2025b.
  75. Wentong Li, Yuqian Yuan, Jian Liu, Dongqi Tang, Song Wang, Jie Qin, Jianke Zhu, and Lei Zhang. Tokenpacker: Efficient visual projector for multimodal llm. In IJCV, 2025c. find
  76. Xuying Li, Zhuo Li, Yuji Kosuga, and Victor Bian. Output length effect on deepseek-r1’s safety in forced thinking. arXiv preprint arXiv:2503.01923, 2025d. — [arXiv:2503.01923](https://arxiv.org/abs/2503.01923) find
  77. Yiwei Li, Peiwen Yuan, Shaoxiong Feng, Boyuan Pan, Bin Sun, Xinglin Wang, Heda Wang, and Kan Li. Turning dust into gold: Distilling complex reasoning capabilities from llms by leveraging negative data. In AAAI, 2024a. find
  78. Yiwei Li, Peiwen Yuan, Shaoxiong Feng, Boyuan Pan, Xinglin Wang, Bin Sun, Heda Wang, and Kan Li. Escape sky-high cost: Early-stopping self-consistency for multi-step reasoning. arXiv preprint arXiv:2401.10480, 2024b. find
  79. Yuetai Li, Xiang Yue, Zhangchen Xu, Fengqing Jiang, Luyao Niu, Bill Yuchen Lin, Bhaskar Ramasubramanian, and Radha Poovendran. Small models struggle to learn from strong reasoners. arXiv preprint arXiv:2502.12143, 2025e. find
  80. Yun Li, Lin Niu, Xipeng Zhang, Kai Liu, Jianchen Zhu, and Zhanhui Kang. E-sparse: Boosting the large language model inference through entropy-based n: M sparsity. arXiv preprint arXiv:2310.15929, 2023b. find
  81. Baohao Liao, Yuhui Xu, Hanze Dong, Junnan Li, Christof Monz, Silvio Savarese, Doyen Sahoo, and Caiming Xiong. Reward-guided speculative decoding for efficient llm reasoning. arXiv preprint arXiv:2501.19324, 2025a. — [arXiv:2501.19324](https://arxiv.org/abs/2501.19324) find
  82. Huanxuan Liao, Shizhu He, Yupu Hao, Xiang Li, Yuanzhe Zhang, Jun Zhao, and Kang Liu. Skintern: Internalizing symbolic knowledge for distilling better cot capabilities into small language models. In COLING, 2025b.
  83. Jonathan Light, Wei Cheng, Wu Yue, Masafumi Oyamada, Mengdi Wang, Santiago Paternain, and Haifeng Chen. Disc: Dynamic decomposition improves llm inference scaling. arXiv preprint arXiv:2502.16706, 2025. — [arXiv:2502.16706](https://arxiv.org/abs/2502.16706) find
  84. 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 ICLR, 2023. — Let's Verify Step by Step
  85. Ji Lin, Jiaming Tang, Haotian Tang, Shang Yang, Wei-Ming Chen, Wei-Chen Wang, Guangxuan Xiao, Xingyu Dang, Chuang Gan, and Song Han. Awq: Activation-aware weight quantization for llm compression and acceleration. In MLSys, 2024. find
  86. Wang Ling, Dani Yogatama, Chris Dyer, and Phil Blunsom. Program induction by rationale generation: Learning to solve and explain algebraic word problems. arXiv preprint arXiv:1705.04146, 2017. find
  87. Fan Liu, Wenshuo Chao, Naiqiang Tan, and Hao Liu. Bag of tricks for inference-time computation of llm reasoning. arXiv preprint arXiv:2502.07191, 2025a. find
  88. Jinyi Liu, Yan Zheng, Rong Cheng, Qiyu Wu, Wei Guo, Fei Ni, Hebin Liang, Yifu Yuan, Hangyu Mao, Fuzheng Zhang, et al. From chaos to order: The atomic reasoner framework for fine-grained reasoning in large language models. arXiv preprint arXiv:2503.15944, 2025b. — [arXiv:2503.15944](https://arxiv.org/abs/2503.15944) find
  89. Junnan Liu, Hongwei Liu, Linchen Xiao, Ziyi Wang, Kuikun Liu, Songyang Gao, Wenwei Zhang, Songyang Zhang, and Kai Chen. Are your llms capable of stable reasoning? arXiv preprint arXiv:2412.13147, 2024a. find
  90. Ruikang Liu, Yuxuan Sun, Manyi Zhang, Haoli Bai, Xianzhi Yu, Tiezheng Yu, Chun Yuan, and Lu Hou. Quantization hurts reasoning? an empirical study on quantized reasoning models. arXiv preprint arXiv:2504.04823, 2025c. — [arXiv:2504.04823](https://arxiv.org/abs/2504.04823) find
  91. Runze Liu, Junqi Gao, Jian Zhao, Kaiyan Zhang, Xiu Li, Biqing Qi, Wanli Ouyang, and Bowen Zhou. Can 1b llm surpass 405b llm? rethinking compute-optimal test-time scaling. arXiv preprint arXiv:2502.06703, 2025d.
  92. Tengxiao Liu, Qipeng Guo, Xiangkun Hu, Cheng Jiayang, Yue Zhang, Xipeng Qiu, and Zheng Zhang. Can language models learn to skip steps? arXiv preprint arXiv:2411.01855, 2024b. — [arXiv:2411.01855](https://arxiv.org/abs/2411.01855) find
  93. Tianqiao Liu, Zui Chen, Zitao Liu, Mi Tian, and Weiqi Luo. Expediting and elevating large language model reasoning via hidden chain-of-thought decoding. arXiv preprint arXiv:2409.08561, 2024c. — [arXiv:2409.08561](https://arxiv.org/abs/2409.08561) find
  94. Yufan Liu, Jiajiong Cao, Bing Li, Chunfeng Yuan, Weiming Hu, Yangxi Li, and Yunqiang Duan. Knowledge distillation via instance relationship graph. In CVPR, 2019. find
  95. Enzhe Lu, Zhejun Jiang, Jingyuan Liu, Yulun Du, Tao Jiang, Chao Hong, Shaowei Liu, Weiran He, Enming Yuan, Yuzhi Wang, et al. Moba: Mixture of block attention for long-context llms. arXiv preprint arXiv:2502.13189, 2025.
  96. Haipeng Luo, Qingfeng Sun, Can Xu, Pu Zhao, Jianguang Lou, Chongyang Tao, Xiubo Geng, Qingwei Lin, Shifeng Chen, and Dongmei Zhang. Wizardmath: Empowering mathematical reasoning for large language models via reinforced evol-instruct. arXiv preprint arXiv:2308.09583, 2023.
  97. 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. arXiv preprint arXiv:2501.12570, 2025a. find
  98. Michael Luo, Sijun Tan, Justin Wong, Xiaoxiang Shi, William Y Tang, Manan Roongta, Colin Cai, Jeffrey Luo, Tianjun Zhang, Li Erran Li, et al. Deepscaler: Surpassing o1-preview with a 1.5 b model by scaling rl. Notion Blog, 2025b. find
  99. Yijia Luo, Yulin Song, Xingyao Zhang, Jiaheng Liu, Weixun Wang, GengRu Chen, Wenbo Su, and Bo Zheng. Deconstructing long chain-of-thought: A structured reasoning optimization framework for long cot distillation. arXiv preprint arXiv:2503.16385, 2025c. find
  100. Chang Ma, Haiteng Zhao, Junlei Zhang, Junxian He, and Lingpeng Kong. Non-myopic generation of language models for reasoning and planning. arXiv preprint arXiv:2410.17195, 2024. — [arXiv:2410.17195](https://arxiv.org/abs/2410.17195) find
  101. Xinyin Ma, Gongfan Fang, and Xinchao Wang. Llm-pruner: On the structural pruning of large language models. In NeurIPS, 2023.
  102. Xinyin Ma, Guangnian Wan, Runpeng Yu, Gongfan Fang, and Xinchao Wang. compressible chain-of-thought tuning. arXiv preprint arXiv:2502.09601, 2025. — [arXiv:2502.09601](https://arxiv.org/abs/2502.09601) find
  103. Cot-valve: Length-Aman Madaan, Niket Tandon, Prakhar Gupta, Skyler Hallinan, Luyu Gao, Sarah Wiegreffe, Uri Alon, Nouha Dziri, Shrimai Prabhumoye, Yiming Yang, et al. Self-refine: Iterative refinement with self-feedback. In NeurIPS, 2023. find
  104. Lucie Charlotte Magister, Jonathan Mallinson, Jakub Adamek, Eric Malmi, and Aliaksei Severyn. Teaching small language models to reason. arXiv preprint arXiv:2212.08410, 2022.
  105. Ethan Mendes and Alan Ritter. Language models can self-improve at state-value estimation for better search. arXiv preprint arXiv:2503.02878, 2025. — [arXiv:2503.02878](https://arxiv.org/abs/2503.02878) find
  106. Niklas Muennighoff, Zitong Yang, Weijia Shi, Xiang Lisa Li, Li Fei-Fei, Hannaneh Hajishirzi, Luke Zettlemoyer, Percy Liang, Emmanuel Candès, and Tatsunori Hashimoto. s1: Simple test-time scaling. arXiv preprint arXiv:2501.19393, 2025.
  107. Tergel Munkhbat, Namgyu Ho, Seohyun Kim, Yongjin Yang, Yujin Kim, and Se-Young Yun. Self-training elicits concise reasoning in large language models. arXiv preprint arXiv:2502.20122, 2025. find
  108. Xuefei Ning, Zinan Lin, Zixuan Zhou, Zifu Wang, Huazhong Yang, and Yu Wang. Skeleton-of-thought: Prompting llms for efficient parallel generation. arXiv preprint arXiv:2307.15337, 2023. find
  109. Isaac Ong, Amjad Almahairi, Vincent Wu, Wei-Lin Chiang, Tianhao Wu, Joseph E Gonzalez, M Waleed Kadous, and Ion Stoica. Routellm: Learning to route llms with preference data. arXiv preprint arXiv:2406.18665, 2024. find
  110. OpenAI. OpenAI o1. https://openai.com/o1/, 2024. ↗
  111. Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, et al. Training language models to follow instructions with human feedback. In NeurIPS, 2022.
  112. Shuyi Ouyang, Hongyi Wang, Shiao Xie, Ziwei Niu, Ruofeng Tong, Yen-Wei Chen, and Lanfen Lin. Slvit: Scale-wise language-guided vision transformer for referring image segmentation. In IJCAI, 2023. find
  113. Daniele Paliotta, Junxiong Wang, Matteo Pagliardini, Kevin Y Li, Aviv Bick, J Zico Kolter, Albert Gu, François Fleuret, and Tri Dao. Thinking slow, fast: Scaling inference compute with distilled reasoners. arXiv preprint arXiv:2502.20339, 2025. — [arXiv:2502.20339](https://arxiv.org/abs/2502.20339) find
  114. Rui Pan, Yinwei Dai, Zhihao Zhang, Gabriele Oliaro, Zhihao Jia, and Ravi Netravali. Specreason: Fast and accurate inference-time compute via speculative reasoning. arXiv preprint arXiv:2504.07891, 2025. — [arXiv:2504.07891](https://arxiv.org/abs/2504.07891) find
  115. Shubham Parashar, Blake Olson, Sambhav Khurana, Eric Li, Hongyi Ling, James Caverlee, and Shuiwang Ji. Inference-time computations for llm reasoning and planning: A benchmark and insights. arXiv preprint arXiv:2502.12521, 2025. — [arXiv:2502.12521](https://arxiv.org/abs/2502.12521) find
  116. Jacob Pfau, William Merrill, and Samuel R Bowman. Let’s think dot by dot: Hidden computation in transformer language models. In COLM, 2024. find
  117. S Joe Qin and Thomas A Badgwell. An overview of industrial model predictive control technology. In AIche symposium series, 1997. find
  118. Xiaoye Qu, Yafu Li, Zhaochen Su, Weigao Sun, Jianhao Yan, Dongrui Liu, Ganqu Cui, Daizong Liu, Shuxian Liang, Junxian He, et al. A survey of efficient reasoning for large reasoning models: Language, multimodality, and beyond. arXiv preprint arXiv:2503.21614, 2025a.
  119. Yuxiao Qu, Matthew YR Yang, Amrith Setlur, Lewis Tunstall, Edward Emanuel Beeching, Ruslan Salakhutdinov, and Aviral Kumar. Optimizing test-time compute via meta reinforcement fine-tuning. arXiv preprint arXiv:2503.07572, 2025b. find
  120. Matthew Renze and Erhan Guven. The benefits of a concise chain of thought on problem-solving in large language models. In FLLM, 2024. find
  121. Abulhair Saparov and He He. Language models are greedy reasoners: A systematic formal analysis of chain-of-thought. In ICLR, 2023.
  122. Nikunj Saunshi, Nishanth Dikkala, Zhiyuan Li, Sanjiv Kumar, and Sashank J Reddi. Reasoning with latent thoughts: On the power of looped transformers. In ICLR, 2025. find
  123. Victor Schmidt, Kamal Goyal, Aditya Joshi, Boris Feld, Liam Conell, Nikolas Laskaris, Doug Blank, Jonathan Wilson, Sorelle Friedler, and Sasha Luccioni. Codecarbon: estimate and track carbon emissions from machine learning computing (2021). DOI: https://doi.org/10.5281/zenodo, 4658424, 2021. — [doi:10.5281/zenodo](https://doi.org/10.5281/zenodo) ↗
  124. Kele Shao, Keda Tao, Kejia Zhang, Sicheng Feng, Mu Cai, Yuzhang Shang, Haoxuan You, Can Qin, Yang Sui, and Huan Wang. When tokens talk too much: A survey of multimodal long-context token compression across images, videos, and audios. arXiv preprint arXiv:2507.20198, 2025. — [arXiv:2507.20198](https://arxiv.org/abs/2507.20198) find
  125. Xuan Shen, Yizhou Wang, Xiangxi Shi, Yanzhi Wang, Pu Zhao, and Jiuxiang Gu. Efficient reasoning with hidden thinking. arXiv preprint arXiv:2501.19201, 2025a. — [arXiv:2501.19201](https://arxiv.org/abs/2501.19201) find
  126. 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. arXiv preprint arXiv:2503.04472, 2025b. find
  127. Zhenyi Shen, Hanqi Yan, Linhai Zhang, Zhanghao Hu, Yali Du, and Yulan He. Codi: Compressing chain-of-thought into continuous space via self-distillation. arXiv preprint arXiv:2502.21074, 2025c.
  128. Charlie Snell, Jaehoon Lee, Kelvin Xu, and Aviral Kumar. Scaling llm test-time compute optimally can be more effective than scaling model parameters. arXiv preprint arXiv:2408.03314, 2024.
  129. Mingyang Song, Mao Zheng, Zheng Li, Wenjie Yang, Xuan Luo, Yue Pan, and Feng Zhang. Fastcurl: Curriculum reinforcement learning with progressive context extension for efficient training r1-like reasoning models. arXiv preprint arXiv:2503.17287, 2025. — [arXiv:2503.17287](https://arxiv.org/abs/2503.17287) find
  130. Zayne Sprague, Fangcong Yin, Juan Diego Rodriguez, Dongwei Jiang, Manya Wadhwa, Prasann Singhal, Xinyu Zhao, Xi Ye, Kyle Mahowald, and Greg Durrett. To cot or not to cot? chain-of-thought helps mainly on math and symbolic reasoning. arXiv preprint arXiv:2409.12183, 2024.
  131. Gaurav Srivastava, Shuxiang Cao, and Xuan Wang. Towards reasoning ability of small language models. arXiv preprint arXiv:2502.11569, 2025.
  132. DiJia Su, Hanlin Zhu, Yingchen Xu, Jiantao Jiao, Yuandong Tian, and Qinqing Zheng. Token assorted: Mixing latent and text tokens for improved language model reasoning. arXiv preprint arXiv:2502.03275, 2025. — [arXiv:2502.03275](https://arxiv.org/abs/2502.03275) find
  133. 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, 2025a.
  134. Yang Sui, Yu-Neng Chuang, Guanchu Wang, Jiamu Zhang, Tianyi Zhang, Jiayi Yuan, Hongyi Liu, Andrew Wen, Shaochen Zhong, Hanjie Chen, and Xia Hu. Stop overthinking: A survey on efficient reasoning for large language models. arXiv preprint arXiv:2503.16419, 2025b.
  135. Yuan Sui, Yufei He, Tri Cao, Simeng Han, and Bryan Hooi. Meta-reasoner: Dynamic guidance for optimized inference-time reasoning in large language models. arXiv preprint arXiv:2502.19918, 2025c.
  136. Hanshi Sun, Momin Haider, Ruiqi Zhang, Huitao Yang, Jiahao Qiu, Ming Yin, Mengdi Wang, Peter Bartlett, and Andrea Zanette. Fast best-of-n decoding via speculative rejection. In NeurIPS, 2024a. find
  137. Yutao Sun, Hangbo Bao, Wenhui Wang, Zhiliang Peng, Li Dong, Shaohan Huang, Jianyong Wang, and Furu Wei. Multimodal latent language modeling with next-token diffusion. arXiv preprint arXiv:2412.08635, 2024b. — [arXiv:2412.08635](https://arxiv.org/abs/2412.08635) find
  138. Richard S Sutton. Learning to predict by the methods of temporal differences. Machine learning, 1988. find
  139. Wenhui Tan, Jiaze Li, Jianzhong Ju, Zhenbo Luo, Jian Luan, and Ruihua Song. Think silently, think fast: Dynamic latent compression of llm reasoning chains. arXiv preprint arXiv:2505.16552, 2025.
  140. Amir Taubenfeld, Tom Sheffer, Eran Ofek, Amir Feder, Ariel Goldstein, Zorik Gekhman, and Gal Yona. Confidence improves self-consistency in llms. arXiv preprint arXiv:2502.06233, 2025. find
  141. Kimi Team, Angang Du, Bofei Gao, Bowei Xing, Changjiu Jiang, Cheng Chen, Cheng Li, Chenjun Xiao, Chenzhuang Du, Chonghua Liao, et al. Kimi k1. 5: Scaling reinforcement learning with llms. arXiv preprint arXiv:2501.12599, 2025.
  142. Fengwei Teng, Zhaoyang Yu, Quan Shi, Jiayi Zhang, Chenglin Wu, and Yuyu Luo. Atom of thoughts for markov llm test-time scaling. arXiv preprint arXiv:2502.12018, 2025. — [arXiv:2502.12018](https://arxiv.org/abs/2502.12018) find
  143. Kaiwen Tuo and Huan Wang. Sparsessm: Efficient selective structured state space models can be pruned in one-shot. arXiv preprint arXiv:2506.09613, 2025. — [arXiv:2506.09613](https://arxiv.org/abs/2506.09613) find
  144. Karthik Valmeekam, Matthew Marquez, Sarath Sreedharan, and Subbarao Kambhampati. On the planning abilities of large language models-a critical investigation. In NeurIPS, 2023. find
  145. Aaron Van Den Oord, Oriol Vinyals, et al. Neural discrete representation learning. In NeurIPS, 2017. find
  146. Guangya Wan, Yuqi Wu, Jie Chen, and Sheng Li. Reasoning aware self-consistency: Leveraging reasoning paths for efficient llm sampling. arXiv preprint arXiv:2408.17017, 2024.
  147. Ante Wang, Linfeng Song, Ye Tian, Dian Yu, Haitao Mi, Xiangyu Duan, Zhaopeng Tu, Jinsong Su, and Dong Yu. Don’t get lost in the trees: Streamlining llm reasoning by overcoming tree search exploration pitfalls. arXiv preprint arXiv:2502.11183, 2025a. find
  148. Huan Wang, Can Qin, Yulun Zhang, and Yun Fu. Neural pruning via growing regularization. In ICLR, 2021. find
  149. Junxiong Wang, Wen-Ding Li, Daniele Paliotta, Daniel Ritter, Alexander M Rush, and Tri Dao. M1: Towards scalable test-time compute with mamba reasoning models. arXiv preprint arXiv:2504.10449, 2025b. — [arXiv:2504.10449](https://arxiv.org/abs/2504.10449) find
  150. Lei Wang, Chen Ma, Xueyang Feng, Zeyu Zhang, Hao Yang, Jingsen Zhang, Zhiyuan Chen, Jiakai Tang, Xu Chen, Yankai Lin, et al. A survey on large language model based autonomous agents. Frontiers of Computer Science, 18(6):186345, 2024a.
  151. Song Wang, Gongfan Fang, Lingdong Kong, Xiangtai Li, Jianyun Xu, Sheng Yang, Qiang Li, Jianke Zhu, and Xinchao Wang. Pixelthink: Towards efficient chain-of-pixel reasoning. arXiv preprint arXiv:2505.23727, 2025c. — [arXiv:2505.23727](https://arxiv.org/abs/2505.23727) find
  152. Xinglin Wang, Shaoxiong Feng, Yiwei Li, Peiwen Yuan, Yueqi Zhang, Chuyi Tan, Boyuan Pan, Yao Hu, and Kan Li. Make every penny count: Difficulty-adaptive self-consistency for cost-efficient reasoning. arXiv preprint arXiv:2408.13457, 2024b. find
  153. Xinyi Wang, Lucas Caccia, Oleksiy Ostapenko, Xingdi Yuan, William Yang Wang, and Alessandro Sordoni. Guiding language model reasoning with planning tokens. In COLM, 2024c. find
  154. Xuezhi Wang, Jason Wei, Dale Schuurmans, Quoc Le, Ed Chi, Sharan Narang, Aakanksha Chowdhery, and Denny Zhou. Self-consistency improves chain of thought reasoning in language models. arXiv preprint arXiv:2203.11171, 2022a.
  155. Yiming Wang, Pei Zhang, Siyuan Huang, Baosong Yang, Zhuosheng Zhang, Fei Huang, and Rui Wang. Sampling-efficient test-time scaling: Self-estimating the best-of-n sampling in early decoding. arXiv preprint arXiv:2503.01422, 2025d. find
  156. Yizhong Wang, Yeganeh Kordi, Swaroop Mishra, Alisa Liu, Noah A Smith, Daniel Khashabi, and Hannaneh Hajishirzi. Self-instruct: Aligning language models with self-generated instructions. arXiv preprint arXiv:2212.10560, 2022b. find
  157. Yue Wang, Qiuzhi Liu, Jiahao Xu, Tian Liang, Xingyu Chen, Zhiwei He, Linfeng Song, Dian Yu, Juntao Li, Zhuosheng Zhang, et al. Thoughts are all over the place: On the underthinking of o1-like llms. arXiv preprint arXiv:2501.18585, 2025e. find
  158. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, et al. Chain-of-thought prompting elicits reasoning in large language models. In NeurIPS, 2022.
  159. Liang Wen, Yunke Cai, Fenrui Xiao, Xin He, Qi An, Zhenyu Duan, Yimin Du, Junchen Liu, Lifu Tang, Xiaowei Lv, et al. Light-r1: Curriculum sft, dpo and rl for long cot from scratch and beyond. arXiv preprint arXiv:2503.10460, 2025. find
  160. Han Wu, Yuxuan Yao, Shuqi Liu, Zehua Liu, Xiaojin Fu, Xiongwei Han, Xing Li, Hui-Ling Zhen, Tao Zhong, and Mingxuan Yuan. Unlocking efficient long-to-short llm reasoning with model merging. arXiv preprint arXiv:2503.20641, 2025a. — [arXiv:2503.20641](https://arxiv.org/abs/2503.20641) find
  161. Siye Wu, Jian Xie, Yikai Zhang, Aili Chen, Kai Zhang, Yu Su, and Yanghua Xiao. Arm: Adaptive reasoning model. arXiv preprint arXiv:2505.20258, 2025b. — [arXiv:2505.20258](https://arxiv.org/abs/2505.20258) find
  162. Yangzhen Wu, Zhiqing Sun, Shanda Li, Sean Welleck, and Yiming Yang. Inference scaling laws: An empirical analysis of compute-optimal inference for problem-solving with language models. In ICLR, 2025c.
  163. Yuyang Wu, Yifei Wang, Tianqi Du, Stefanie Jegelka, and Yisen Wang. When more is less: Understanding chain-of-thought length in llms. arXiv preprint arXiv:2502.07266, 2025d. find
  164. Heming Xia, Yongqi Li, Chak Tou Leong, Wenjie Wang, and Wenjie Li. Tokenskip: Controllable chain-of-thought compression in llms. arXiv preprint arXiv:2502.12067, 2025.
  165. Kun Xiang, Zhili Liu, Zihao Jiang, Yunshuang Nie, Kaixin Cai, Yiyang Yin, Runhui Huang, Haoxiang Fan, Hanhui Li, Weiran Huang, Yihan Zeng, Yu-Jie Yuan, Jianhua Han, Lanqing Hong, Hang Xu, and Xiaodan Liang. Can atomic step decomposition enhance the self-structured reasoning of multimodal large model… — [arXiv:2503.06252](https://arxiv.org/abs/2503.06252) find
  166. Kun Xiang, Zhili Liu, Zihao Jiang, Yunshuang Nie, Kaixin Cai, Yiyang Yin, Runhui Huang, Haoxiang Fan, Hanhui Li, Weiran Huang, et al. Can atomic step decomposition enhance the self-structured reasoning of multimodal large models? arXiv preprint arXiv:2503.06252, 2025b. — [arXiv:2503.06252](https://arxiv.org/abs/2503.06252) find
  167. Guangxuan Xiao, Ji Lin, Mickael Seznec, Hao Wu, Julien Demouth, and Song Han. SmoothQuant: Accurate and efficient post-training quantization for large language models. In ICML, 2023. find
  168. Fangzhi Xu, Hang Yan, Chang Ma, Haiteng Zhao, Jun Liu, Qika Lin, and Zhiyong Wu. ϕ-decoding: Adaptive foresight sampling for balanced inference-time exploration and exploitation. arXiv preprint arXiv:2503.13288, 2025a. — [arXiv:2503.13288](https://arxiv.org/abs/2503.13288) find
  169. Fengli Xu, Qianyue Hao, Zefang Zong, Jingwei Wang, Yunke Zhang, Jingyi Wang, Xiaochong Lan, Jiahui Gong, Tianjian Ouyang, Fanjin Meng, et al. Towards large reasoning models: A survey of reinforced reasoning with large language models. arXiv preprint arXiv:2501.09686, 2025b. find
  170. Silei Xu, Wenhao Xie, Lingxiao Zhao, and Pengcheng He. Chain of draft: Thinking faster by writing less. arXiv preprint arXiv:2502.18600, 2025c. find
  171. Yige Xu, Xu Guo, Zhiwei Zeng, and Chunyan Miao. Softcot: Soft chain-of-thought for efficient reasoning with llms. arXiv preprint arXiv:2502.12134, 2025d.
  172. Yuchen Yan, Yongliang Shen, Yang Liu, Jin Jiang, Mengdi Zhang, Jian Shao, and Yueting Zhuang. Inftythink: Breaking the length limits of long-context reasoning in large language models. arXiv preprint arXiv:2503.06692, 2025.
  173. An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, et al. Qwen2. 5 technical report. arXiv preprint arXiv:2412.15115, 2024a. find
  174. Chenxiao Yang, Nathan Srebro, David McAllester, and Zhiyuan Li. Pencil: Long thoughts with short memory. arXiv preprint arXiv:2503.14337, 2025a. — [arXiv:2503.14337](https://arxiv.org/abs/2503.14337) find
  175. Enneng Yang, Li Shen, Guibing Guo, Xingwei Wang, Xiaochun Cao, Jie Zhang, and Dacheng Tao. Model merging in llms, mllms, and beyond: Methods, theories, applications and opportunities. arXiv preprint arXiv:2408.07666, 2024b. find
  176. Junjie Yang, Ke Lin, and Xing Yu. Think when you need: Self-adaptive chain-of-thought learning. arXiv preprint arXiv:2504.03234, 2025b. — [arXiv:2504.03234](https://arxiv.org/abs/2504.03234) find
  177. Wen Yang, Minpeng Liao, and Kai Fan. Markov chain of thought for efficient mathematical reasoning. arXiv preprint arXiv:2410.17635, 2024c. find
  178. Wenkai Yang, Shuming Ma, Yankai Lin, and Furu Wei. Towards thinking-optimal scaling of test-time compute for llm reasoning. arXiv preprint arXiv:2502.18080, 2025c.
  179. Zhilin Yang, Peng Qi, Saizheng Zhang, Yoshua Bengio, William W Cohen, Ruslan Salakhutdinov, and Christopher D Manning. Hotpotqa: A dataset for diverse, explainable multi-hop question answering. arXiv preprint arXiv:1809.09600, 2018.
  180. Shunyu Yao, Dian Yu, Jeffrey Zhao, Izhak Shafran, Tom Griffiths, Yuan Cao, and Karthik Narasimhan. Tree of thoughts: Deliberate problem solving with large language models. In NeurIPS, 2023. find
  181. Shunyu Yao, Noah Shinn, Pedram Razavi, and Karthik Narasimhan. τ -bench: A benchmark for tool-agent-user interaction in real-world domains. arXiv preprint arXiv:2406.12045, 2024.
  182. Yixin Ye, Zhen Huang, Yang Xiao, Ethan Chern, Shijie Xia, and Pengfei Liu. Limo: Less is more for reasoning. arXiv preprint arXiv:2502.03387, 2025.
  183. Longhui Yu, Weisen Jiang, Han Shi, Jincheng Yu, Zhengying Liu, Yu Zhang, James T Kwok, Zhenguo Li, Adrian Weller, and Weiyang Liu. Metamath: Bootstrap your own mathematical questions for large language models. arXiv preprint arXiv:2309.12284, 2023.
  184. Ping Yu, Jing Xu, Jason Weston, and Ilia Kulikov. Distilling system 2 into system 1. arXiv preprint arXiv:2407.06023, 2024. find
  185. Qifan Yu, Zhenyu He, Sijie Li, Xun Zhou, Jun Zhang, Jingjing Xu, and Di He. Enhancing auto-regressive chain-of-thought through loop-aligned reasoning. arXiv preprint arXiv:2502.08482, 2025a.
  186. 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, 2025b.
  187. Jingyang Yuan, Huazuo Gao, Damai Dai, Junyu Luo, Liang Zhao, Zhengyan Zhang, Zhenda Xie, YX Wei, Lean Wang, Zhiping Xiao, et al. Native sparse attention: Hardware-aligned and natively trainable sparse attention. arXiv preprint arXiv:2502.11089, 2025.
  188. Weihao Zeng, Yuzhen Huang, Qian Liu, Wei Liu, Keqing He, Zejun Ma, and Junxian He. Simplerl-zoo: Investigating and taming zero reinforcement learning for open base models in the wild. arXiv preprint arXiv:2503.18892, 2025a.
  189. Zhiyuan Zeng, Qinyuan Cheng, Zhangyue Yin, Yunhua Zhou, and Xipeng Qiu. Revisiting the test-time scaling of o1-like models: Do they truly possess test-time scaling capabilities? arXiv preprint arXiv:2502.12215, 2025b.
  190. Jintian Zhang, Yuqi Zhu, Mengshu Sun, Yujie Luo, Shuofei Qiao, Lun Du, Da Zheng, Huajun Chen, and Ningyu Zhang. Lightthinker: Thinking step-by-step compression. arXiv preprint arXiv:2502.15589, 2025a.
  191. Nan Zhang, Yusen Zhang, Prasenjit Mitra, and Rui Zhang. When reasoning meets compression: Benchmarking compressed large reasoning models on complex reasoning tasks. arXiv preprint arXiv:2504.02010, 2025b. — [arXiv:2504.02010](https://arxiv.org/abs/2504.02010) find
  192. Yulun Zhang, Huan Wang, Can Qin, and Yun Fu. Learning efficient image super-resolution networks via structure-regularized pruning. In ICLR, 2021. find
  193. Yunxiang Zhang, Muhammad Khalifa, Lajanugen Logeswaran, Jaekyeom Kim, Moontae Lee, Honglak Lee, and Lu Wang. Small language models need strong verifiers to self-correct reasoning. arXiv preprint arXiv:2404.17140, 2024. find
  194. Yichun Zhao, Shuheng Zhou, and Huijia Zhu. Probe then retrieve and reason: Distilling probing and reasoning capabilities into smaller language models. In LREC-COLING, 2024. find
  195. Huaixiu Steven Zheng, Swaroop Mishra, Hugh Zhang, Xinyun Chen, Minmin Chen, Azade Nova, Le Hou, Heng-Tze Cheng, Quoc V Le, Ed H Chi, et al. Natural plan: Benchmarking llms on natural language planning. arXiv preprint arXiv:2406.04520, 2024. find
  196. Denny Zhou, Nathanael Schärli, Le Hou, Jason Wei, Nathan Scales, Xuezhi Wang, Dale Schuurmans, Claire Cui, Olivier Bousquet, Quoc Le, et al. Least-to-most prompting enables complex reasoning in large language models. In ICLR, 2023.
  197. Zhi Zhou, Tan Yuhao, Zenan Li, Yuan Yao, Lan-Zhe Guo, Xiaoxing Ma, and Yu-Feng Li. Bridging internal probability and self-consistency for effective and efficient llm reasoning. arXiv preprint arXiv:2502.00511, 2025. — [arXiv:2502.00511](https://arxiv.org/abs/2502.00511) find
  198. Jiace Zhu, Yingtao Shen, Jie Zhao, and An Zou. Path-consistency: Prefix enhancement for efficient inference in llm. arXiv preprint arXiv:2409.01281, 2024a. — [arXiv:2409.01281](https://arxiv.org/abs/2409.01281) find
  199. Xunyu Zhu, Jian Li, Can Ma, and Weiping Wang. Improving mathematical reasoning capabilities of small language models via feedback-driven distillation. arXiv preprint arXiv:2411.14698, 2024b. — [arXiv:2411.14698](https://arxiv.org/abs/2411.14698) find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/feng-2025-efficient-reasoning-models-survey/