On-Policy Distillation

Follow this field Ask about this field

Recent momentum

emerging

0 papers in the last 28 days · 0.0% of indexed attention

Twelve weeks of publication activity for this topic as it is defined today.

Weekly history

Recent digests

What was published in this field, kept on the site without email delivery.

Period ending 2026-09-21

14 new papers

A weekly snapshot of new work published in On-Policy Distillation.

Period ending 2026-09-14

20 new papers

A weekly snapshot of new work published in On-Policy Distillation.

Period ending 2026-09-07

10 new papers

A weekly snapshot of new work published in On-Policy Distillation.

Inside this field

Focused directions

Focused topic

On-Policy Distillation

Within On-Policy Distillation

291 papersExplore

Focused topic

Unsupervised On-Policy Self-Distillation

Within On-Policy Distillation

116 papersExplore

Focused topic

Teacher

Within On-Policy Distillation

81 papersExplore

Focused topic

Rp-Opsd

Within On-Policy Distillation

52 papersExplore

Focused topic

Token-Level Supervision

Within On-Policy Distillation

44 papersExplore

Focused topic

Teacher-Student Log-Probability Gap

Within On-Policy Distillation

19 papersExplore

437 papers

Latest in On-Policy Distillation

Browse all topics →

Hidden Thoughts Are Not Secret: Reasoning Trace Exposure in LLMs
May 30, 2026Yu-An Lu, Ci-Yang Tsai, Yu-Lin Tsai +2Reasoning Traces Large Language Model Reasoning
Decomposed On-Policy Distillation for Vision-Language Reasoning: Steering Gradients for Visual Grounding
May 30, 2026Hee Suk Yoon, Eunseop Yoon, Jaehyun Jang +6Vision-Language On-Policy Distillation
Interpretable Policy Distillation for Power Grid Topology Control
May 30, 2026Aleksandra Dmitruka, Karlis FreivaldsProximal Policy Optimization Power Systems
Weak Critics Make Strong Learners: On-Policy Critique Distillation for Scalable Oversight
May 29, 2026Can Jin, Jiakang Li, Rui Wu +2Weak Supervision Critic-Free Reinforcement Learning
Bridging Reasoning Trajectories in On-Policy Distillation via Near-Future Guidance
May 29, 2026Yuxuan Jiang, Francis FerraroOn-Policy Distillation Reasoning Trajectory
Are Full Rollouts Necessary for On-Policy Distillation?
May 29, 2026Yaocheng Zhang, Jiajun Chai, Yuqian Fu +7On-Policy Distillation Long Rollouts
CAST: Non-Privileged Clipped Asymmetric Self-Teaching with Advantage Flipping for GRPO
May 29, 2026Yang Li, Gongle Xue, Yijia Guo +3Reinforcement Learning With Verifiable Reward Unsupervised On-Policy Self-Distillation
Trust-Region Behavior Blending for On-Policy Distillation
May 29, 2026Daniil Plyusov, Alexey Gorbatovski, Alexey Malakhov +4On-Policy Distillation Trust Region
Inference-Free Multimodal Learned Sparse Retrieval for Production-Scale Visual Document Search
May 29, 2026Gyu-Hwung Cho, Youngjune Lee, Kiyoon Jeong +5Visual Document Retrieval Multimodal Retrieval
Distilling LLM Feedback for Lean Theorem Proving
May 29, 2026Gaetan Narozniak, Gérard Biau, Rémi Munos +2Token-Level Supervision Group Relative Policy Optimization
Your Teacher Can't Help You Here: Combating Supervision Fidelity Decay in On-Policy Distillation
May 29, 2026Yanjiang Liu, Jie Lou, Xinyan Guan +7On-Policy Distillation Teacher
Same Evidence, Different Answers: Canonical-Context On-Policy Distillation for Multi-Turn Language Models
May 28, 2026Zizhuo Lin, Quanling Liu, Jinsheng Quan +6On-Policy Distillation Math Problems
A Predictive Law for On-Policy Self-Distillation From World Feedback
May 28, 2026Tommy He, Jerome Sieber, Matteo SaponatiUnsupervised On-Policy Self-Distillation On-Policy Distillation
GAPD: Gold-Action Policy Distillation for Agentic Reinforcement Learning in Knowledge Base Question Answering
May 28, 2026Xin Sun, Jianan Xie, Zhongqi Chen +6Agentic Reinforcement-Learning Framework Deep Reinforcement Learning
GDSD: Reinforcement Learning as Guided Denoiser Self-Distillation for Diffusion Language Models
May 28, 2026Xiaohang Tang, Keyue Jiang, Che Liu +4Diffusion Language Models Unsupervised On-Policy Self-Distillation
Draft-OPD: On-Policy Distillation for Speculative Draft Models
May 28, 2026Haodi Lei, Yafu Li, Haoran Zhang +8Speculative Decoding Self-Speculative
Surfacing Isolated Learners with Outcome-Independent Mediation of Feedback between Teachers and Students Using AI
May 28, 2026Junsoo Park, Youssef Medhat, Htet Phyo Wai +2Learners Feedback Learning
OISD: On-Policy Internal Self-Distillation of Language Models
May 27, 2026Xinyu Liu, Darryl Cherian Jacob, Yang Zhou +2Unsupervised On-Policy Self-Distillation On-Policy Distillation
Skill-Conditioned Gated Self-Distillation for LLM Reasoning
May 27, 2026Jiazhen Huang, Xiao Chen, Xiao Luo +3Unsupervised On-Policy Self-Distillation Large Language Model Reasoning
ADWIN: Adaptive Windows for Horizon-Aware On-Policy Distillation
May 27, 2026Kun Liang, Chenming Tang, Clive Bai +3On-Policy Distillation Adaptive Testing
Teacher-Student Representational Alignment for Reinforcement Learning-Driven Imitation Learning
May 27, 2026Meraj Mammadov, Pedro Zuidberg Dos Martires, Johannes Andreas StorkImitation Learning Imitation
From Fact Overwriting to Knowledge Evolution: Causal Editing via On-Policy Self-Distillation
May 27, 2026Shuaike Li, Kai Zhang, Xianquan Wang +2Multimodal Knowledge Editing Knowledge Distillation
Data-Efficient On-Policy Distillation for Automatic Speech Recognition
May 27, 2026Yu Lin, Yiming Wang, Runyuan Cai +1Llm-Based Automatic Speech Recognition Automatic Speech Recognition Systems
ROSD: Reflective On-Policy Self-Distillation for Language Model Reasoning across Domains
May 27, 2026Ziqi Zhao, Xinyu Ma, Liu Yang +6Unsupervised On-Policy Self-Distillation Large Language Model Reasoning
Reflective Dialogue between Teacher and Solver Agents for Video Question Answering
May 27, 2026Takuya Murakawa, Toru TamakiVideo Question Answering In-Context Learning
Restoring the Sweet Spot: Pass-Rate Weighted Self-Distillation for LLM Reasoning
May 26, 2026Zehao Liu, Yuanpu Cao, Jinghui Chen +1Unsupervised On-Policy Self-Distillation Batch Distillation
Not All Tokens Matter Equally: Dynamic In-context Vector Distillation with Decisive-Token Supervision for Long-form Medical Report Generation
May 26, 2026Ning Wu, Rui Liu, Xinkun Lin +5Medical Report Generation Token-Level Supervision
MAIGO: Mitigating Lost-in-Conversation with History-Cleaned On-Policy Self-Distillation
May 26, 2026Haoyu Zheng, Yun Zhu, Shu Yuan +5Unsupervised On-Policy Self-Distillation
StepOPSD: Step-Aware Online Preference Distillation for Agent Reinforcement Learning
May 26, 2026Yanfei Zhang, Xu Lin, Chenglin WuOn-Policy Distillation Rp-Opsd
Counteraction-Aware Multi-Teacher On-Policy Distillation for General Capability Recovery with Domain Preservation
May 26, 2026Tianlei Chen, Jiao Ou, Ziyuan Liu +3On-Policy Distillation Large Language Model Post-Training
Adversarial Dual On-Policy Distillation from Expressive Teacher
May 26, 2026Zhenglin Wan, Jingxuan Wu, Xingrui Yu +5On-Policy Distillation Behavior Cloning
Less is More: Early Stopping Rollout for On-Policy Distillation
May 26, 2026Zhou Ziheng, Jiaqi Li, Huacong Tang +2On-Policy Distillation Off-Policy Evaluation
When the Strongest Teacher Is Not the Best Teacher: Student-Centric Answer Selection
May 26, 2026Zhengyu Hu, Zheyuan Xiao, Linxin Song +10Teacher Training Data
Not All Disagreement Is Learnable: Token Teachability in On-Policy Distillation
May 26, 2026Yuanyi Wang, Su Lu, Yanggan Gu +6On-Policy Distillation Token-Level Supervision
DV-SFT: Direct Vision Supervision for Fine-Grained Visual Understanding
May 26, 2026Jianfei Zhao, Feng Zhang, Xin Sun +3Visual Tokens Fine-Grained Perception
Dynamic Neural Koopman Distillation for Real-Time Robot Control Using Diffusion Models
May 24, 2026Lei Zheng, Peiqi Yu, Zengqi Peng +2Diffusion Policies Robot Planning
Strong Teacher Not Needed? On Distillation in LLM Pretraining
May 22, 2026Taiming Lu, Zhuang LiuChain-Of-Thought Distillation Teacher-Student Log-Probability Gap
EDGE-OPD: Internalizing Privileged Context with Evidence Guided On-Policy Distillation
May 22, 2026Aristotelis Lazaridis, Dylan Bates, Aman Sharma +3On-Policy Distillation Privilege Illusion
The Distillation Game: Adaptive Attacks & Efficient Defenses
May 21, 2026Youssef Allouah, Mahdi Haghifam, Sanmi Koyejo +1Teacher
Post-Training is About States, Not Tokens: A State Distribution View of SFT, RL, and On-Policy Distillation
May 21, 2026Dong NieLLM Post-Training Methods Supervised Fine-Tuning
Self-Policy Distillation via Capability-Selective Subspace Projection
May 21, 2026Guangya Hao, Yitong Shang, Yunbo Long +2Unsupervised On-Policy Self-Distillation On-Policy Distillation
Search-E1: Self-Distillation Drives Self-Evolution in Search-Augmented Reasoning
May 21, 2026Zihan Liang, Yufei Ma, Ben Chen +4Search-Augmented Reasoning Unsupervised On-Policy Self-Distillation
A Boundary-Layer Mechanism for One-Third Scaling in Online Softmax Classification
May 21, 2026Marcel Kühn, Yoon Thelge, Bernd RosenowLearning Rates Scaling Laws
Tailoring Teaching to Aptitude: Direction-Adaptive Self-Distillation for LLM Reasoning
May 21, 2026Hongbin Zhang, Chaozheng Wang, Kehai Chen +4Unsupervised On-Policy Self-Distillation On-Policy Distillation
TextTeacher: What Can Language Teach About Images?
May 21, 2026Tobias Christian Nauen, Stanislav Frolov, Brian Bernhard Moser +3Imagenet Knowledge Distillation
Toward Understanding Adversarial Distillation: Why Robust Teachers Fail
May 21, 2026Hongsin Lee, Hye Won ChungMulti-Teacher Distillation Chain-Of-Thought Distillation
Visual-Advantage On-Policy Distillation for Vision-Language Models
May 21, 2026Ruiqi Liu, Xiaolei Lv, Gengsheng Li +8Large Vision Language Models On-Policy Distillation
OPPO: Bayesian Value Recursion for Token-Level Credit Assignment in LLM Reasoning
May 21, 2026Yu Li, Rui Miao, Tian Lan +1Large Language Model Reasoning Credit Assignment
X-Token: Projection-Guided Cross-Tokenizer Knowledge Distillation
May 20, 2026Sharath Turuvekere Sreenivas, Adithyakrishna Venkatesh Hanasoge, Mingyu Yang +4Multi-Teacher Distillation Teacher
When Are Teacher Tokens Reliable? Position-Weighted On-Policy Self-Distillation for Reasoning
May 20, 2026Xiaogeng Liu, Xinyan Wang, Yingzi Ma +2On-Policy Distillation Teacher
Towards Understanding Self-Pretraining for Sequence Classification
May 20, 2026Omar Coser, Loredana Zollo, Paolo Soda +1Self-Supervised Transformer Architectures
AVSD: Adaptive-View Self-Distillation by Balancing Consensus and Teacher-Specific Privileged Signals
May 20, 2026Duy Nguyen, Hanqi Xiao, Archiki Prasad +7Teacher Inference-Time
Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation
May 18, 2026Qianhao Yuan, Jie Lou, Xing Yu +4Fine-Grained Perception On-Policy Distillation
AMR-SD: Asymmetric Meta-Reflective Self-Distillation for Token-Level Credit Assignment
May 18, 2026Zhenlin Wei, Pu Jian, Yingzhuo Deng +6Unsupervised On-Policy Self-Distillation Token-Level Supervision
SD-Search: On-Policy Hindsight Self-Distillation for Search-Augmented Reasoning
May 18, 2026Yufei Ma, Zihan Liang, Ben Chen +6Search-Augmented Reasoning Unsupervised On-Policy Self-Distillation
HINT-SD: Targeted Hindsight Self-Distillation for Long-Horizon Agents
May 18, 2026Woongyeng Yeo, Yumin Choi, Taekyung Ki +1Long-Horizon Agents Hindsight
$\boldsymbol{f}$ -OPD: Stabilizing Long-Horizon On-Policy Distillation with Freshness-Aware Control
May 18, 2026Xianwei Chen, Shimin Zhang, Jibin WuOn-Policy Distillation Rp-Opsd
MoASE++: Mixture of Activation Sparsity Experts with Domain-Adaptive On-policy Distillation for Continual Test Time Adaptation
May 18, 2026Ronyu Zhang, Aosong Cheng, Gaole Dai +8Continual Test-Time Adaptation Model Pretraining
Self-Supervised On-Policy Distillation for Reasoning Language Models
May 17, 2026Zhiquan Tan, Yinrong HongUnsupervised On-Policy Self-Distillation On-Policy Distillation
Decoupling KL and Trajectories: A Unified Perspective for SFT, DAgger, Offline RL, and OPD in LLM Distillation
May 16, 2026Anhao Zhao, Haoran Xin, Yingqi Fan +3On-Policy Distillation Knowledge Distillation

May 30, 2026 · cs.AIJ/K move · Enter open · S save

Hidden Thoughts Are Not Secret: Reasoning Trace Exposure in LLMs

Yu-An Lu, Ci-Yang Tsai, Yu-Lin Tsai, Raluca Ada Popa+1

National Yang Ming Chiao Tung University · UC Berkeley

Reasoning traces have become a valuable form of learning signals for improving and transferring the capabilities of large language models. In particular, detailed traces can help distill reasoning behavior from stronger teacher models into weaker student models. The value of capability transfer has motivated many deployed systems with reasoning models to hide raw internal traces and expose at most summaries and answers to users. As a result, we ask whether such interface-level trace hiding prevents users from obtaining useful reasoning supervision through prompting. We study this question with Reasoning Exposure Prompting (REP), a lightweight in-context elicitation method that uses shadow-model-generated demonstrations wrapped in auxiliary code-like formats to raise user-visible reasoning traces from a victim model. Across the common reasoning dataset, different victim models, and different student model distillation, REP substantially increases similarity between exposed and REP-conditioned internal traces while preserving useful reasoning signals.

Reasoning Traces Large Language Model Reasoning Latent Thoughts Teacher

Open paper Ask about it PDF