{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,2]],"date-time":"2026-05-02T14:56:26Z","timestamp":1777733786380,"version":"3.51.4"},"reference-count":79,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"11","license":[{"start":{"date-parts":[[2025,11,1]],"date-time":"2025-11-01T00:00:00Z","timestamp":1761955200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,11,1]],"date-time":"2025-11-01T00:00:00Z","timestamp":1761955200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,11,1]],"date-time":"2025-11-01T00:00:00Z","timestamp":1761955200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"National Science Fund for Distinguished Young Scholars of China","award":["62325604"],"award-info":[{"award-number":["62325604"]}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62276271"],"award-info":[{"award-number":["62276271"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62421002"],"award-info":[{"award-number":["62421002"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Neural Netw. Learning Syst."],"published-print":{"date-parts":[[2025,11]]},"DOI":"10.1109\/tnnls.2025.3591266","type":"journal-article","created":{"date-parts":[[2025,8,18]],"date-time":"2025-08-18T19:45:16Z","timestamp":1755546316000},"page":"19820-19833","source":"Crossref","is-referenced-by-count":2,"title":["Distilling Reasoning Ability From Large Language Models With Adaptive Thinking"],"prefix":"10.1109","volume":"36","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0704-4543","authenticated-orcid":false,"given":"Xiaoshu","family":"Chen","sequence":"first","affiliation":[{"name":"College of Computer Science and Technology, National University of Defense Technology, Changsha, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1491-4594","authenticated-orcid":false,"given":"Sihang","family":"Zhou","sequence":"additional","affiliation":[{"name":"College of Intelligence Science and Technology, National University of Defense Technology, Changsha, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4837-455X","authenticated-orcid":false,"given":"Ke","family":"Liang","sequence":"additional","affiliation":[{"name":"College of Computer Science and Technology, National University of Defense Technology, Changsha, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9066-1475","authenticated-orcid":false,"given":"Xinwang","family":"Liu","sequence":"additional","affiliation":[{"name":"College of Computer Science and Technology, National University of Defense Technology, Changsha, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Navigate through enigmatic labyrinth a survey of chain of thought reasoning: Advances, frontiers and future","author":"Chu","year":"2023","journal-title":"arXiv:2309.15402"},{"key":"ref2","first-page":"1877","article-title":"Language models are few-shot learners","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Brown"},{"key":"ref3","article-title":"LLaMA: Open and efficient foundation language models","author":"Touvron","year":"2023","journal-title":"arXiv:2302.13971"},{"key":"ref4","article-title":"Llama 2: Open foundation and fine-tuned chat models","author":"Touvron","year":"2023","journal-title":"arXiv:2307.09288"},{"key":"ref5","first-page":"24824","article-title":"Chain-of-Thought prompting elicits reasoning in large language models","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Lee"},{"key":"ref6","first-page":"14852","article-title":"Large language models are reasoning teachers","volume-title":"Proc. 61st Annu. Meeting Assoc. Comput. Linguistics","author":"Ho"},{"key":"ref7","first-page":"8003","article-title":"Distilling step-by-step! Outperforming larger language models with less training data and smaller model sizes","volume-title":"Proc. Findings Assoc. Comput. Linguistics: ACL","author":"Hsieh"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v38i17.29821"},{"key":"ref9","first-page":"2195","article-title":"Specializing pre-trained language models for better relational reasoning via network pruning","volume-title":"Proc. Findings Assoc. Comput. Linguistics, NAACL","author":"Ren"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.naacl-long.142"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2022.3220933"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/TKDE.2024.3392335"},{"key":"ref13","article-title":"A survey of knowledge graph reasoning on graph types: Static, dynamic, and multimodal","author":"Liang","year":"2022","journal-title":"arXiv:2212.05767"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2021.3070843"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1007\/s11432-024-4222-0"},{"key":"ref16","article-title":"Teaching small language models to reason","author":"Charlotte Magister","year":"2022","journal-title":"arXiv:2212.08410"},{"key":"ref17","volume-title":"Thinking, Fast Slow","author":"Kahneman","year":"2011"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1111\/ejn.70024"},{"key":"ref19","first-page":"10764","article-title":"PAL: Program-aided language models","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Gao"},{"key":"ref20","article-title":"Program of thoughts prompting: Disentangling computation from reasoning for numerical reasoning tasks","author":"Chen","year":"Oct. 2022","journal-title":"Trans. Mach. Learn. Res."},{"key":"ref21","first-page":"37","article-title":"MathPrompter: Mathematical reasoning using large language models","volume-title":"Proc. 61st Annu. Meeting Assoc. Comput. Linguistics","author":"Imani"},{"key":"ref22","article-title":"Dynamic prompt learning via policy gradient for semi-structured mathematical reasoning","volume-title":"Proc. Int. Conf. Learn. Represent. (ICLR)","author":"Lu"},{"key":"ref23","article-title":"Automatic chain of thought prompting in large language models","volume-title":"Proc. 11th Int. Conf. Learn. Represent. (ICLR)","author":"Zhang"},{"key":"ref24","article-title":"Large language model guided tree-of-thought","author":"Long","year":"2023","journal-title":"arXiv:2305.08291"},{"key":"ref25","first-page":"11809","article-title":"Tree of thoughts: Deliberate problem solving with large language models","volume-title":"Proc. 37th Int. Conf. Neural Inf. Process. Syst. (NIPS)","author":"Yao","year":"2023"},{"key":"ref26","article-title":"Boosting logical reasoning in large language models through a new framework: The graph of thought","author":"Lei","year":"2023","journal-title":"arXiv:2308.08614"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v38i16.29720"},{"key":"ref28","article-title":"Chain-of-Dictionary prompting elicits translation in large language models","author":"Lu","year":"2023","journal-title":"arXiv:2305.06575"},{"key":"ref29","article-title":"Knowledge-driven CoT: Exploring faithful reasoning in LLMs for knowledge-intensive question answering","author":"Wang","year":"2023","journal-title":"arXiv:2308.13259"},{"key":"ref30","article-title":"Self-consistency improves chain of thought reasoning in language models","author":"Wang","year":"2022","journal-title":"arXiv:2203.11171"},{"key":"ref31","article-title":"Training verifiers to solve math word problems","author":"Cobbe","year":"2021","journal-title":"arXiv:2110.14168"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1503.02531"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1145\/3459637.3481909"},{"key":"ref34","first-page":"5776","article-title":"MiniLM: Deep self-attention distillation for task-agnostic compression of pre-trained transformers","volume-title":"Proc. 34th Int. Conf. Neural Inf. Process. Syst.","author":"Wang"},{"key":"ref35","first-page":"4323","article-title":"Patient knowledge distillation for BERT model compression","volume-title":"Proc. Conf. Empirical Methods Natural Lang. Process. 9th Int. Joint Conf. Natural Lang. Process. (EMNLP-IJCNLP)","author":"Sun"},{"key":"ref36","first-page":"6748","article-title":"Mind\u2019s mirror: Distilling self-evaluation capability and comprehensive thinking from large language models","volume-title":"Proc. Conf. North Amer. Chapter Assoc. Comput. Linguistics, Human Lang. Technol.","author":"Liu"},{"key":"ref37","first-page":"11307","article-title":"MinT: Boosting generalization in mathematical reasoning via multi-view fine-tuning","volume-title":"Proc. Joint Int. Conf. Comput. Linguistics","author":"Liang"},{"key":"ref38","first-page":"6857","article-title":"Learning to maximize mutual information for chain-of-thought distillation","volume-title":"Proc. Findings Assoc. Comput. Linguistics","author":"Chen"},{"key":"ref39","first-page":"10421","article-title":"Specializing smaller language models towards multi-step reasoning","volume-title":"Proc. 40th Int. Conf. Mach. Learn.","author":"Fu"},{"key":"ref40","first-page":"1673","article-title":"Mixed distillation helps smaller language models reason better","volume-title":"Proc. Findings Assoc. Comput. Linguistics, EMNLP","author":"Chenglin"},{"key":"ref41","article-title":"UnicoTT: A unified framework for structural chain-of-thought distillation","volume-title":"Proc. The 13th Int. Conf. Learn. Represent.","author":"Zhuang"},{"key":"ref42","first-page":"5514","article-title":"DISCO: Distilling counterfactuals with large language models","volume-title":"Proc. 61st Annu. Meeting Assoc. for Comput. Linguistics","author":"Chen"},{"key":"ref43","first-page":"5546","article-title":"SCOTT: Self-consistent chain-of-thought distillation","volume-title":"Proc. 61st Annu. Meeting Assoc. for Comput. Linguistics","author":"Wang"},{"key":"ref44","first-page":"5831","article-title":"Teaching small language models reasoning through counterfactual distillation","volume-title":"Proc. Conf. Empirical Methods Natural Lang. Process.","author":"Feng"},{"key":"ref45","first-page":"9074","article-title":"Enhancing the reasoning capabilities of small language models via solution guidance fine-tuning","volume-title":"Proc. 31st Int. Conf. Comput. Linguistics","author":"Bi"},{"key":"ref46","first-page":"13241","article-title":"Keypoint-based progressive Chain-of-Thought distillation for LLMs","volume-title":"Proc. 41st Int. Conf. Mach. Learn.","author":"Feng"},{"key":"ref47","first-page":"7059","article-title":"Distilling reasoning capabilities into smaller language models","volume-title":"Proc. Findings Assoc. Comput. Linguistics","author":"Shridhar"},{"key":"ref48","article-title":"Long-short Chain-of-Thought mixture supervised fine-tuning eliciting efficient reasoning in large language models","author":"Yu","year":"2025","journal-title":"arXiv:2505.03469"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v39i23.34608"},{"key":"ref50","article-title":"ThinkSwitcher: When to think hard, when to think fast","author":"Liang","year":"2025","journal-title":"arXiv:2505.14183"},{"key":"ref51","article-title":"HDFlow: Enhancing LLM complex problem-solving with hybrid thinking and dynamic workflows","author":"Yao","year":"2024","journal-title":"arXiv:2409.17433"},{"key":"ref52","article-title":"Dualformer: Controllable fast and slow thinking by learning with randomized reasoning traces","author":"Su","year":"2024","journal-title":"arXiv:2410.09918"},{"issue":"8","key":"ref53","first-page":"9","article-title":"Language models are unsupervised multitask learners","volume":"1","author":"Radford","year":"2019","journal-title":"OpenAI blog"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1016\/j.aiopen.2023.08.012"},{"key":"ref55","first-page":"4582","article-title":"Prefix-tuning: Optimizing continuous prompts for generation","volume-title":"Proc. 59th Annu. Meeting Assoc. Comput. Linguistics 11th Int. Joint Conf. Natural Lang. Process.","author":"Li"},{"key":"ref56","first-page":"3045","article-title":"The power of scale for parameter-efficient prompt tuning","volume-title":"Proc. Conf. Empirical Methods Natural Lang. Process.","author":"Lester"},{"key":"ref57","first-page":"5039","article-title":"SPoT: Better frozen model adaptation through soft prompt transfer","volume-title":"Proc. 60th Annu. Meeting Assoc. Comput. Linguistics","author":"Vu"},{"key":"ref58","first-page":"3949","article-title":"On transferability of prompt tuning for natural language processing","volume-title":"Proc. Conf. North Amer. Chapter Assoc. Comput. Linguistics, Human Lang. Technol.","author":"Su"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.acl-long.381"},{"issue":"1","key":"ref60","first-page":"5485","article-title":"Exploring the limits of transfer learning with a unified text-to-text transformer","volume":"21","author":"Raffel","year":"2019","journal-title":"J. Mach. Learn. Res."},{"key":"ref61","doi-asserted-by":"crossref","DOI":"10.1109\/NNICE64954.2025.11063722","article-title":"Knowledge fusion of large language models","volume-title":"Proc. 12th Int. Conf. Learn. Represent.","author":"Wan"},{"key":"ref62","article-title":"When less is more: Investigating data pruning for pretraining LLMs at scale","volume-title":"Proc. NeurIPS Workshop Attributing Model Behav. Scale","author":"Marion"},{"key":"ref63","first-page":"585","article-title":"Parsing algebraic word problems into equations","volume":"3","author":"Koncel-Kedziorski","year":"2015","journal-title":"Trans. Annu. Meet. Assoc. Comput. Linguist."},{"key":"ref64","first-page":"523","article-title":"Learning to solve arithmetic word problems with verb categorization","volume-title":"Proc. Conf. Empirical Methods Natural Lang. Process. (EMNLP)","author":"Hosseini"},{"key":"ref65","first-page":"1743","article-title":"Solving general arithmetic word problems","volume-title":"Proc. Conf. Empirical Methods Natural Lang. Process","author":"Roy"},{"key":"ref66","first-page":"158","article-title":"Program induction by rationale generation: Learning to solve and explain algebraic word problems","volume-title":"Proc. 55th Annu. Meeting Assoc. Comput. Linguistics","author":"Ling"},{"key":"ref67","first-page":"2080","article-title":"Are NLP models really able to solve simple math word problems?","volume-title":"Proc. Conf. North Amer. Chapter Assoc. Comput. Linguistics, Human Lang. Technol.","author":"Patel"},{"key":"ref68","first-page":"22199","article-title":"Large language models are zero-shot reasoners","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"35","author":"Kojima"},{"key":"ref69","first-page":"4149","article-title":"CommonsenseQA: A question answering challenge targeting commonsense knowledge","volume-title":"Proc. Conf. North American Chapter Assoc. Comput. Linguistics, Human Lang. Technol.","author":"Talmor"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00370"},{"key":"ref71","first-page":"27730","article-title":"Training language models to follow instructions with human feedback","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Ouyang"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.2307\/j.ctvcm4g18.8"},{"key":"ref73","article-title":"ChatGLM: A family of large language models from GLM-130B to GLM-4 all tools","author":"Glm","year":"2024","journal-title":"arXiv:2406.12793"},{"key":"ref74","article-title":"Qwen2 technical report","volume-title":"arXiv:2407.10671","author":"Yang","year":"2024"},{"key":"ref75","article-title":"The llama 3 herd of models","author":"Grattafiori","year":"2024","journal-title":"arXiv:2407.21783"},{"key":"ref76","article-title":"Chain of draft: Thinking faster by writing less","author":"Xu","year":"2025","journal-title":"arXiv:2502.18600"},{"key":"ref77","first-page":"46534","article-title":"Self-refine: Iterative refinement with self-feedback","volume-title":"Proc. 37th Conf. Neural Inf. Process. Syst.","author":"Madaan"},{"key":"ref78","first-page":"17889","article-title":"Encouraging divergent thinking in large language models through multi-agent debate","volume-title":"Proc. Conf. Empirical Methods Natural Lang. Process.","author":"Liang"},{"key":"ref79","article-title":"CMATH: Can your language model pass Chinese elementary school math test?","author":"Wei","year":"2023","journal-title":"arXiv:2306.16636"}],"container-title":["IEEE Transactions on Neural Networks and Learning Systems"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/5962385\/11220834\/11127195.pdf?arnumber=11127195","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,30]],"date-time":"2025-10-30T18:03:21Z","timestamp":1761847401000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11127195\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,11]]},"references-count":79,"journal-issue":{"issue":"11"},"URL":"https:\/\/doi.org\/10.1109\/tnnls.2025.3591266","relation":{},"ISSN":["2162-237X","2162-2388"],"issn-type":[{"value":"2162-237X","type":"print"},{"value":"2162-2388","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,11]]}}}