The paper introduces On-Policy Mid-Training Distillation (OPMD), a method that shifts continual pre‑training from passive next‑token prediction to active, query‑driven knowledge acquisition. By sampling trajectories from the evolving student policy and applying dense token‑level reverse‑KL rewards from a frozen teacher, OPMD trains the model on states it actually visits, reducing the mismatch between training and generation. Experiments on MATH‑500 and Minerva show that OPMD improves a Qwen student’s performance over its base and outperforms the standard next‑token prediction baseline, while integrating seamlessly into existing continual pre‑training pipelines without inference‑time overhead.
By Ruize Zheng, Xuyang Liu, Siyu Ren, Xiaobing Tu, Yinggui Wang, Jinkui Ren, Xiantao Zhang, Linfeng Zhang
arXiv:2606. 06712v1 Announce Type: cross Abstract: We study the transformation of autoregressive models (ARLMs) into diffusion language models (DLMs).
By Xingyu Su, Jacob Helwig, Shubham Parashar, Atharv Chagi, Lakshmi Jotsna, Degui Zhi, James Caverlee, Dileep Kalathil, Shuiwang Ji
arXiv:2610.19601v1 Announce Type: new
Abstract: Knowledge distillation (KD) improves language-model pretraining by matching teacher distributions at corpus prefixes, but provides no direct supervisio...
By Zhenhe Wu, Jian Yang, Yaping Jin, Xinlai Kang, Jie Hu, Hanting Chen
arXiv:2606. 09456v1 Announce Type: new Abstract: On-Policy Distillation (OPD) has become a core technique in the post-training of Large Language Models (LLMs) for transferring knowledge from domain experts to student models.
By Yifan Niu, Han Xiao, Dongyi Liu, Zelong Wang, Dihong Gong, Yasheng Wang, Jia Li
Knowledge distillation (KD) improves language-model pretraining by matching teacher distributions at corpus prefixes, but provides no direct supervision at contexts induced by student predictions. We...
arXiv:2608.28647v1 Announce Type: new
Abstract: Target-only post-training can improve performance in a specialized domain while degrading behaviors that a general-purpose base model acquired before a...
By Yifei Li, Rongman Xu, Lingling Zhang, Muye Huang, Zihan Ma, Jiashuai Liu, Hang Yan, Heng Wang
arXiv:2605. 03677v2 Announce Type: replace Abstract: On-policy distillation (OPD) has recently emerged as an effective post-training paradigm for consolidating the capabilities of specialized expert models into a single student model.
By Wenjin Hou, Shangpin Peng, Weinong Wang, Zheng Ruan, Yue Zhang, Zhenglin Zhou, Mingqi Gao, Yifei Chen, Kaiqi Wang, Hongming Yang, Chengquan Zhang, Zhuotao Tian, Han Hu, Yi Yang, Fei Wu, Hehe Fan
The paper investigates data efficiency and selection in On‑Policy Distillation (OPD) for large language models. It shows that 1‑shot OPD—training on a single example—consistently improves performance, especially when the example is hard, and that longer chain‑of‑thought (CoT) paths drive the gains rather than token entropy. Based on these findings, the authors propose a simple hard‑example selection strategy that, using only eight carefully chosen hard examples, matches the performance of a 17,000‑example baseline across models from 1.5B to 7B parameters.
By Zhinan Hou, Jiaqi Zhang, Xunliang Cai, Keyou You
Activation-Conditioned Self-Distillation (ACSD) is a new on‑policy self‑distillation method that uses a frozen copy of the base model to extract a steering vector by contrasting activations from self‑generated trajectories that reach verified correct answers with all other trajectories. The student learns from next‑token distributions on its own prefixes, without needing reference text or teacher parameter updates, and is used alone at inference. Across five models, ACSD achieves the highest mean accuracy on four mathematical benchmarks, with notable gains on DeepSeek‑R1‑0528‑Qwen3‑8B and LiveCodeBench v6 compared to the OPSD baseline.
By Zhexi Lu, Subhajit Chaudhury, Tejaswini Pedapati, Keerthiram Murugesan, Lei Yu
The paper introduces vOPD, an On‑Policy Distillation method that incorporates a control variate baseline from reinforcement learning to reduce gradient variance. By using the per‑token negative reverse KL divergence as a closed‑form value function, vOPD maintains a lightweight single‑sample estimator while keeping gradients unbiased. Experiments on mathematics, science, and instruction‑following tasks show vOPD consistently outperforms standard OPD and matches the performance of more expensive full‑vocabulary methods.
By Minjae Oh, Sangjun Song, Gyubin Choi, Yunho Choi, Yohan Jo
arXiv:2608.23493v1 Announce Type: new
Abstract: Self-reflection is a powerful mechanism for credit assignment in human learning, converting sparse outcome feedback into actionable guidance. However,...
By Jialong Liu, Yuling Shi, Ning Yang, Xiaodong Gu, Zuchao Li
arXiv:2609.01532v1 Announce Type: new
Abstract: Logit-based knowledge distillation (KD) is used to train smaller language models (LMs) via supervision from stronger teachers, but whether its benefits...
By Jacqueline He, Howard Yen, Shuyue Stella Li, Margaret Li, Hanqing Zeng, Yinglong Xia, Benyu Zhang, Zhuokai Zhao, Qiang Zhang, Pang Wei Koh, Luke Zettlemoyer, Wen-tau Yih