The paper investigates how on‑policy distillation (OPD) inherits a teacher’s behavior in vision‑language models, comparing it to filtered supervised fine‑tuning (SFT) across different data budgets and tasks. Experiments on ChartQA and Geometry3K show that OPD performs well only when the teacher remains reliable on student‑generated trajectories; otherwise it underperforms SFT and can erase previously learned completion behavior. Filtering teacher demonstrations for completeness largely restores SFT’s advantages, indicating that OPD can transfer both knowledge and persistent failure modes from the teacher.
By Frank F. Yang, Kefei Yan, Yiren Wang, Yiyang Wang
The paper introduces Advantage-Based On-Policy Distillation (AOPD), a method that selectively weights teacher actions during online policy distillation by their counterfactual advantage to the student. Unlike standard OPD, which follows the teacher at every state, AOPD reinforces only those teacher actions that improve the student's expected outcome, using an estimator that relies solely on standard student rollouts and terminal rewards. Experiments on math, coding, and Medqa tasks demonstrate that AOPD outperforms both distillation and reinforcement learning baselines in external-teacher and self-distillation settings.
By Mahdi Sabbaghi, George Pappas, Hamed Hassani, Adel Javanmard
arXiv:2608. 01263v1 Announce Type: new Abstract: On-policy distillation (OPD) samples trajectories from the current student policy and minimizes token-level divergence between student and teacher next-token distributions at prefixes along those trajectories.
By Leyan Xue, Feng Xiong, Mingjun Ma, Changqing Zhang
arXiv:2610.28586v1 Announce Type: cross
Abstract: Agentic on-policy distillation (OPD) trains a student agent on its own trajectories while a teacher scores every token. A natural explanation of wher...
By Shu Zhao, Tan Yu
The paper investigates on‑policy distillation (OPD), showing that teacher supervision during OPD contains significant noise that grows with teacher size, yet the student policy remains largely unaffected by this noise. It finds that OPD’s gains stem mainly from suppressing low‑log‑probability tokens, a process that can be replicated without a teacher. Building on this insight, the authors propose On‑Policy Self‑Adaptation (OPSA), a supervision‑free method that uses entropy‑adaptive negative advantages to improve performance on several benchmarks, outperforming both the base model and OPD.
By Yi Ding, Ruqi Zhang
arXiv:2609.36546v1 Announce Type: cross
Abstract: On-policy distillation (OPD) trains a student model on its self-generated trajectories with dense token-level teacher feedback. However, naive OPD ma...
By Shutong Wu, Xiwen Chen, Brendan Rappazzo, Daiheng Zhang, Anderson Schneider, Yuriy Nevmyvaka, Jiawei Zhang
The paper introduces Teacher-Gated On-Policy Distillation (TGOPD), a method that verifies teacher reliability at the prompt level before applying dense supervision in on-policy distillation. TGOPD uses verifier-scored teacher probes to decide whether to route a prompt to dense OPD or to a verifier-grounded alternative. Experiments on 4B and 35B models across mathematics, code, and instruction tasks show TGOPD outperforms vanilla OPD and improves teacher GPU utilization from 9.8% to 78.9% in a 4B single-domain run.
By Zhiwei Zhang, Zechen Sun, Fei Zhao, Kang Peng, Bin Liang, Huayu Deng, Yao Hu, Kam-Fai Wong, Mu Chuan
arXiv:2512.22802v2 Announce Type: replace-cross
Abstract: Step distillation accelerates diffusion sampling by training a few-step student to imitate a many-step teacher, but distillation itself remai...
By Amirhossein Tighkhorshid, Zahra Dehghanian, Hamid R. Rabiee
arXiv:2609. 40285v1 Announce Type: new Abstract: On-policy distillation (OPD) is a promising approach for training language agents, providing dense teacher supervision on student-generated trajectories.
By Yinghui He, Yapei Chang, Khushi Bhardwaj, Daniele Molinari, Tugrul Konuk, Jan Kautz, Ali Hatamizadeh
CA-OPD is a confidence‑aware on‑policy distillation framework that improves structured visual prediction by using teacher confidence to selectively correct unreliable student transitions and gradually transfer rollout control to the student. The method aligns supervision with intervention decisions, providing direct cross‑entropy loss for corrected tokens and full predictive distribution for retained tokens. In a multi‑teacher setting for GUI grounding and OCR, CA‑OPD significantly outperforms the Qwen3.5‑0.8B baseline, achieving large gains on benchmarks such as ScreenSpot‑Pro and OCRBench‑v2 English.
By Menghao Li, Linjie Mu, Yin Wang, Haotian Hu, Yannian Gu, Lujiayi Xue, Fanyi Wang
The paper proposes a reward‑density principle for language‑model post‑training: sparse sequence‑level rewards are best for exploring and discovering new behaviors in a strong teacher, while dense token‑level teacher supervision is optimal for compressing that behavior into a smaller deployment model. The authors validate this rule with a four‑stage workflow—teacher RL, forward‑KL warmup, on‑policy distillation, and optional post‑bridge student RL—showing that an RL‑improved 8B teacher distilled through a dense bridge outperforms direct GRPO on a 1.7B student across math and code benchmarks. Ablation studies confirm each stage’s contribution, and the approach generalizes to Llama and Qwen models, offering a practical guideline for allocating scarce labeled data.
By Yuanda Xu, Hejian Sang, Zhengze Zhou, Ran He, Zhipeng Wang, Alborz Geramifard
arXiv:2609.38282v1 Announce Type: new
Abstract: Vision-language models may rewrite anomalous text in images into linguistically plausible expressions, compromising OCR transcription faithfulness. Seq...
By Baode Wang, Zuming Huang, Kexuan Ren, Jun Huang, Wei Chu