Matching Supervision to the Student's Learning Capacity: A Unified Framework for On-Policy Self-Distillation
Yongkang Yang, Zhezheng Hao, Hong Zhang, Yi Liu, Xiankun Lin, Wence Ji, Fanjunduo Wei, Jiarui Yu, Qiang Lin, Xiaoyun Liang, Hande Dong
cs.AI, cs.LG
Submitted: 2026-08-08
Updated: 2026-08-11
Code: https://github.com/lauvlalala/USD
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Training Verifiers to Solve Math Word Problems
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- Adaptive Teacher Exposure for Self-Distillation in LLM Reasoning
- Distilling the Knowledge in a Neural Network
- Uni-OPD: Unifying On-Policy Distillation with a Dual-Perspective Recipe
- Reinforcement Learning via Self-Distillation
- Respecting Self-Uncertainty in On-Policy Self-Distillation for Efficient LLM Reasoning
- Hindsight Compacts but Does Not Repair: Rethinking On-Policy Self-Distillation in Reasoning Models
- Why Does Self-Distillation (Sometimes) Degrade the Reasoning Capability of LLMs?
- DemoPSD: Disagreement-Modulated Policy Self-Distillation
- Filter, Then Reweight: Rethinking Optimization Granularity in On-Policy Distillation
- Rethinking On-Policy Distillation of Large Language Models: Phenomenology, Mechanism, and Recipe
- Prefix Teach, Suffix Fade: Local Teachability Collapse in Strong-to-Weak On-Policy Distillation
- When Are Teacher Tokens Reliable? Position-Weighted On-Policy Self-Distillation for Reasoning
- Training language models to follow instructions with human feedback
- RLCSD: Reinforcement Learning with Contrastive On-Policy Self-Distillation
- Privileged Information Distillation for Language Models
- Purified OPSD: On-Policy Self-Distillation Without Losing How to Think
- Self-Distillation Enables Continual Learning
- A Survey of On-Policy Distillation for Large Language Models
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems