Adaptive Supervised Anchoring for On-Policy Self-Distillation
Meilin Yang, Zixuan Ding, Jianhao Nie, Weite Zhang, Yuxin Zhang, Zhiming Shao, Li Yu, Zhe Fu
Renmin University of China, Beijing, China · Renmin University of China, Beijing, China · Renmin University of China, Beijing, China · Renmin University of China, Beijing, China · Renmin University of China, Beijing, China · Renmin University of China, Beijing, China · Renmin University of China, Beijing, China · Renmin University of China, Beijing, China
cs.LG
Submitted: 2026-08-11
Updated: 2026-08-12
Comments: 9 pages, 4 figures, and 3 tables. Meilin Yang and Zixuan Ding contributed equally. Corresponding authors: Li Yu and Zhe Fu
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- LoRA Learns Less and Forgets Less
- Retaining by Doing: The Role of On-Policy Data in Mitigating Forgetting
- Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
- Distilling the Knowledge in a Neural Network
- The Llama 3 Herd of Models
- An Empirical Study of Catastrophic Forgetting in Large Language Models During Continual Fine-tuning
- GPT-4 Technical Report
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- Self-Distillation Enables Continual Learning
- RL's Razor: Why Online Reinforcement Learning Forgets Less
- ToolAlpaca: Generalized Tool Learning for Language Models with 3000 Simulated Cases
- Qwen3 Technical Report
- OGLS-SD: On-Policy Self-Distillation with Outcome-Guided Logit Steering for LLM Reasoning
- Self-Distilled Reasoner: On-Policy Self-Distillation for Large Language Models
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks