arXiv Machine Learning
On the Generalization Limitation of On-Policy Distillation for Post-RL Consolidation
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv AI
Sep 15
Counteraction-Aware Multi-Teacher On-Policy Distillation for General Capability Recovery with Domain Preservation
The paper introduces Counteraction-Aware Multi-Teacher On-Policy Distillation (CaMOPD), a method designed to recover general capabilities in large language models while preserving domain-specific behavior. CaMOPD tackles two failure modes of standard Multi-Teacher On-Policy Distillation—conflicting recovery and preservation gradients, and weak correction signals—by using decoupled alternating training and selecting samples with large teacher‑student log‑probability gaps. Experiments on role‑play dialogue and medical reasoning QA show that CaMOPD outperforms baselines in general capability recovery while maintaining domain specialization, and gradient coherence analyses confirm more coherent correction signals.
arXiv Machine Learning
Oct 8
Train4Merge: A Controlled Single-Teacher Study of RL vs. SFT Teachers for OPD-Based Model Merging
The study compares supervised fine-tuning (SFT) and reinforcement learning (RL) as teacher training methods for on-policy distillation (OPD) in a single-teacher setting. Using Qwen3.5-9B models across Agentic, Reasoning, and Perception domains, RL-trained teachers consistently produce stronger students and better recovery of performance gains than SFT-trained teachers. The authors attribute this advantage to RL teachers’ smaller parameter shifts from the shared initialization, which facilitates learning through OPD.
arXiv Machine Learning
Aug 13
Towards Understanding On-Policy Distillation through the Lens of Test-Time Scaling
arXiv:2608. 11829v1 Announce Type: new Abstract: On-policy distillation (OPD) has emerged as a promising post-training technique for enhancing LLM reasoning.
arXiv AI
Jul 3
DemoPSD: Disagreement-Modulated Policy Self-Distillation
arXiv:2607. 02502v1 Announce Type: cross Abstract: On-policy self-distillation (OPSD) has emerged as a practical method for training large language models (LLMs) to reason, where a single model acts as both the teacher and the student with different levels of information access.
arXiv Computation and Language
Sep 18
What Does Privileged Information Add to On-Policy Self-Distillation?
The paper investigates how privileged information—such as a teacher’s full solution or reasoning trace—affects on‑policy self‑distillation (OPSD) in language models. Using the AMPLE‑Math benchmark, the authors compare distillation with and without extra teacher views, finding that reference‑free distillation explains most gains for Qwen3‑1.7B, while additional references provide modest benefits, especially for polished solutions. The study also shows that the impact of privileged data depends on the student’s training regime and that altering token‑level supervision can leave student behavior largely unchanged.
arXiv Machine Learning
Oct 2
No Task Vector Is an Island: A Comprehensive Study on the Composability of Task Vectors from On-Policy Distillation
arXiv:2609.39405v1 Announce Type: new Abstract: Task vectors provide a simple mechanism for composing learned capabilities through model merging. However, the composability of task vectors produced b...