CoT Distillation

CoT: Chain-of-Thought

Chain-of-thought distillation trains a student model using a teacher’s intermediate reasoning, aiming to transfer problem-solving behavior as well as correct answers. Common approaches include supervised fine-tuning on teacher-generated traces, matching teacher and student token distributions during student rollouts, and compressing or selecting traces to fit the learner. Methods are tested on mathematical reasoning benchmarks such as GSM8K, MATH, and AIME, as well as code, knowledge, and multimodal tasks. A central tension is that detailed traces can provide useful supervision but may be costly, incorrect, or mismatched to what a student can learn; answer accuracy alone also does not guarantee faithful reasoning.

Latest papers

80 papers13 papers in the last four weeks, up 225% on the four weeks before. 0.1% of all new papers.

From Solver Trajectories to Teaching Trajectories: Cognition-Aligned Reasoning Distillation
Oct 27, 2026Yashuo Luo, Qianren Mao, Ziqi Qin +6Reasoning Distillation Language Model Distillation
National College for Excellent Engineers, Beihang University · Zhongguancun Laboratory · Beihang University +3
TraceForge: Distilling with Your Own Thinking
Oct 21, 2026Yan WangCoT Distillation LLM Reasoning
Independent Researcher
Beyond the Answer: Student-Conditioned Trace Construction for Reasoning Distillation
Oct 16, 2026Shaoang Li, Yanhang Shi, Yufei Li +6CoT Distillation Question Answering
Stony Brook University · Meta AI
Recursive Offline Distillation via Entropy Gain Sampling and Answer Mixing
Oct 14, 2026Andrey Goncharov, Daniil Vyazhev, Simon Epanov +1Online Knowledge Distillation Language Model Distillation
Applied AI Institute
CoTranslator: Reverse to Forward Chain-of-Thought Translator for Reasoning Distillation
Oct 14, 2026Zinan Tang, Zhuoshi Pan, Kairui Chen +5CoT Distillation Language Model Distillation
Tsinghua University · Ant Group · Xiaomi Corporation
Rationale Distribution Distillation via Flow Matching for Distributional Continuous Latent Reasoning
Oct 14, 2026Mingyeong Kim, Gyusang Cho, Juho Lee +1LLM Reasoning Continuous Latent Reasoning
Korea Advanced Institute of Science and Technology
Facts Don't Speak Louder than Words: The Behavioral Essence of Long-CoT Distillation
Oct 13, 2026Yongcan Yu, Jian Liang, Lingxiao He +5CoT Distillation Language Model Distillation
NLPR & MAIS, Institute of Automation, Chinese Academy of Sciences · School of Artificial Intelligence, University of Chinese Academy of Sciences · Meituan +1
Audit Before You Distill: A Controlled Workflow for Reasoning Transfer into Compact Medical Language Models
Oct 12, 2026Tiago Martinelli, João Paulo Papa, Adriano José PereiraLanguage Model Distillation Medical LLMs
Einstein Global Advanced Technologies for Equity (E-GATE), Hospital Israelita Albert Einstein, São Paulo, Brazil · Institute of Science and Technology Itaú (ICTi), Itaú Unibanco, São Paulo, SP, Brazil · School of Sciences, São Paulo State University, São Paulo, Brazil
Overtaught, Underlearned: Learner-Centered Distillation from Multi-Agent Discussion
Oct 12, 2026Mengtao Zhou, Jianxiao Wang, Xiupeng Chen +2Language Model Distillation Knowledge Distillation
The University of Hong Kong · Peking University · Theta Health, Inc.
Learning from Repaired Reasoning: Root-Cause-Guided On-Policy Distillation
Oct 2, 2026Chenglei Shen, Haoyang Yao, Weijie Yu +3Language Model Distillation On-Policy Self-Distillation
Gaoling School of Artificial Intelligence, Renmin University of China · School of Software and Microelectronics, Peking University · School of Information Technology and Management, University of International Business and Economics
Hindsight-Guided Rationale Distillation for Rare Disease Diagnosis
Oct 2, 2026Aarav Singh, Animesh Pathak, Navyansh SinghCoT Distillation Knowledge Distillation
Dr. Shyama Prasad Mukherjee International Institute of Information Technology, Naya Raipur
The Weakest Link: Distilling LLM Reasoning with Worst-Case Constrained Reinforcement Learning
Sep 29, 2026Matthieu Zimmer, Xiaotong Ji, Tu Nguyen +1Language Model Distillation CoT Distillation
Huawei Noah’s Ark Lab · Huawei Heisenberg Research Center · UCL Centre for AI
Learning from Think-Mode Advantage via On-Policy Distillation
Sep 29, 2026Wanqi Ren, Jianxiang Wang, Danxuan Liu +2CoT Distillation Language Model Distillation
ByteDance, China
Distilling Visual Reasoning into Text Space
Sep 28, 2026Wenhan Yang, Nilay Naharas, Ali Payani +1CoT Distillation VLM Distillation
University of California, Los Angeles · Cisco Research
What Shared Prefixes Hide: Trajectory Dropout for On-Policy Distillation
Sep 27, 2026Zizhuo Lin, Quanling Liu, Yi Yang +1CoT Distillation On-Policy Distillation
Zhejiang University
Learning from a Thoughtful Teacher: Adaptive On-Policy Self-Distillation for Mathematical Reasoning
Sep 26, 2026Jiacheng Du, Weiwei Xie, Tianyi Du +3On-Policy Self-Distillation CoT Distillation
National University of Singapore · Shanghai Jiao Tong University · Shanghai Artificial Intelligence Laboratory
Efficient Reasoning Distillation: Small Video-Language Models via Synthetic CoT and Difficulty-Aware Fine-Tuning
Sep 14, 2026Mantek Singh, Jeshwanth Challagundla, Siddharth Raina +1CoT Distillation VLM Distillation
Liverpool John Moores University, Liverpool, England · Carnegie Mellon University, Pittsburgh, USA · Meta, Sunnyvale, USA +1
Aha-Flow Distillation: Flow Markers Matter in LLM Reasoning
Sep 7, 2026Xiaodong Wang, Peixi PengLLM Reasoning CoT Distillation
Peking University · Pengcheng Laboratory
MI-Distillation: Selecting from Model-Interpolated Instruct-Reasoning Data Spectrum for Chain-of-Thought Distillation
Aug 30, 2026Yangsong Lan, Renkai Hu, HongKai Zheng +4CoT Distillation Language Model Distillation
College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing, China · 2The Key Laboratory of Brain-Machine Intelligence Technology, Ministry of Education, Nanjing, China
Beyond Imitation: Auditing the Recoverability of Reasoning in Distilled Models
Aug 11, 2026Ruitong Li, Binjie Guo, Aisheng Mo +4Language Model Distillation Knowledge Distillation
University of Hong Kong · Zhejiang University · Dalian University of Technology +1
Simple-OPD: Demystifying Warm-up for On-policy Distillation
Aug 7, 2026Tao Liu, Taiqiang Wu, Mao Zheng +5CoT Distillation On-Policy Distillation
1Tsinghua University · 2The University of Hong Kong · 3LLM Department, Tencent
CoTinyVLA: Chain-of-Thought Distillation for a Sub-Billion-Parameter Vision-Language-Action Model
Jul 28, 2026Minhyeok Lee, Chiyoung Kim, Chanhoe Gu +5Reasoning for VLA Models Efficient VLA Model Inference
Chung-Ang University, Republic of Korea
EKLAV: Elicited Key CoT Learning via Answer only superVision
Jul 28, 2026Adarsh Singh, Aman Singh, Kushal Raj Bhandari +3Table QA CoT Distillation
Arizona State University · Rensselaer Polytechnic Institute · Scale AI
DAIS: Dependency-Aware Intermediate QA Supervision for Complex Reasoning
Jul 21, 2026Yu Wang, Ming Fan, Xicheng Zhang +5CoT Distillation CoT Supervision
1Xi’an Jiaotong University · 2Huawei Technologies Ltd.
CADENCE: Closing the Reasoning Gap via Coverage-Adaptive On-Policy Distillation
Jul 18, 2026Satyam Kumar, Saurabh JhaCoT Distillation On-Policy Distillation
Better Starts, Better Ends: Bootstrapped Iterative Self-Reasoning Distillation for Compressed Reasoning
Jul 17, 2026Leichao Dong, Dongxu Zhang, Yiding Sun +4CoT Distillation Efficient Language Model Reasoning
1Xi’an Jiaotong University · 2Peking University
Answer-Conditioned Chains of Thought Degrade Verifiable-Reasoning Distillation in Large Language Models
Jul 16, 2026Jungseob Lee, Seungyoon Lee, Suhyune Son +4CoT Distillation Language Model Distillation
1Korea University · 2Zoom Communications · 3Yonsei University
Answer-Conditioned Chain-of-Thought Distillation for Few-Shot Industrial Vision with Small VLMs
Jul 12, 2026Shubham RaoCoT Distillation VLM Adaptation
Entropy AI Research Labs Private Limited
SCOReD: Student-Aware CoT Optimization for Recommendation Distillation
Jul 7, 2026Haz Sameen Shahgir, Yufei Li, Xiaohan Wei +8CoT Distillation Language Model Distillation
University of California Riverside · Meta AI
Purified OPSD: On-Policy Self-Distillation Without Losing How to Think
Jul 2, 2026Zhanming Shen, Jintao Tong, Shaotian Yan +9CoT Distillation On-Policy Self-Distillation
1Zhejiang University · 2Tongyi Lab, Alibaba Group · 3Huazhong University of Science and Technology +1

Oct 27, 2026J/K move · Enter open · S save

From Solver Trajectories to Teaching Trajectories: Cognition-Aligned Reasoning Distillation

Yashuo Luo, Qianren Mao, Ziqi Qin, Junnan Liu+5

National College for Excellent Engineers, Beihang University · Zhongguancun Laboratory · Beihang University+3

Reasoning Distillation Language Model Distillation CoT Distillation RL for Knowledge Distillation

Open paper AskPDF arXiv

Knowledge distillation from large reasoning models (LRMs) often uses raw teacher trajectories as supervision. However, this practice creates a mismatch between the goal of distillation and the nature of teacher trajectories. Advanced LRMs typically improve their reasoning capabilities through reward-driven post-training, such as RLHF. This training optimizes them as strong solvers, rather than as teachers that produce reasoning processes smaller models can easily learn. Since smaller models have different "cognitive capacities" compared to their larger teachers, directly imitating raw teacher trajectories can sometimes be ineffective and often requires a substantial number of high-quality samples. To bridge this mismatch, we propose Cognition-Aligned Reasoning Distillation (CARD), an RL-based framework that transforms solver trajectories into teaching trajectories aligned with the student's cognition. Avoiding expensive resampling, our framework directly transforms existing trajectories. We optimize CARD with two reward signals: (1) a Distribution Alignment Reward that matches the student's predictive distribution, and (2) a Logical Coherence Reward that reduces misleading failed reasoning steps. Comprehensive evaluations across multiple student models and settings demonstrate that CARD outperforms distillation from raw teacher trajectories in in-domain settings while achieving substantial gains in out-of-distribution performance. Code is available at https://github.com/LiteSSLHub/CARD.