MetaOPD introduces a bilevel optimization framework that jointly learns a student model and a lightweight token‑weighting network for on‑policy distillation. The inner objective updates the student using weighted OPD, while the outer objective optimizes the weighting network based on validation loss after a virtual student update, enabling the token‑weight mapping to adapt to the student’s evolving needs. Experiments on multiple reasoning and out‑of‑domain datasets show that MetaOPD improves performance over standard OPD by up to 6.41 percentage points on larger student models.
By Zipeng Wang, Xinpeng Dong, Yuefan Wang, Pingchen Lu, Xian Wei, Kun Kuang, Fei Wu, Zhongxiang Dai, Min Zhang
arXiv:2606. 25488v1 Announce Type: new Abstract: Knowledge Distillation (KD) is widely used to obtain compact models for efficient inference in resource-constrained environments.
By Yifan Wu, Yiqi Wang, Xichen Ye, Wenjing Yan, Xiaoqiang Li, Cheng Jin, Xiangyu Yue, Weizhong Zhang
arXiv:2609.22566v1 Announce Type: cross
Abstract: Knowledge distillation (KD) aims to compress high-performance teacher LLMs into lightweight students. However, distilled students often exhibit subst...
By Dileesha Kannangara, Sanghamitra Dutta
arXiv:2609.13199v1 Announce Type: new
Abstract: Knowledge distillation aims to improve the performance of lightweight student models by transferring knowledge from larger and more powerful teacher mo...
By Dawen Jiang, Zhishu Shen, Zeyu Liu, Tiehua Zhang
The paper introduces TALON, a Task‑Adaptive LoRA‑Teacher framework for Few‑Shot Class‑Incremental Learning. TALON assigns a dedicated LoRA‑Teacher to each incremental task, then distills the frozen teachers into a single LoRA‑Student via Ensemble Knowledge Transfer, using a semantic‑guided weighting scheme to reduce forgetting and overfitting. Experiments on four FSCIL benchmarks show that TALON matches or surpasses state‑of‑the‑art accuracy while using up to 33× fewer deployment parameters and cutting inference time by 41.7%.
By Hongwei Zhao (School of Computer Science,Engineering, Beihang University), Rui Liu (School of Computer Science,Engineering, Beihang University), Yansong Liu (School of Computer Science,Engineering, Beihang University), Zhiyuan Zou (School of Computer Science,Engineering, Beihang University), Yong Chen (School of Computer Science, Beijing University of Posts,Telecommunications)
arXiv:2503. 08038v2 Announce Type: replace-cross Abstract: In this paper, we delve deeper into the Kullback-Leibler (KL) Divergence loss and mathematically prove that it is equivalent to the Decoupled Kullback-Leibler (DKL) Divergence loss that consists of (1) a weighted Mean Square Error (wMSE) loss and (2) a Cross-Entropy loss incorporating soft labels.
By Jiequan Cui, Beier Zhu, Qingshan Xu, Zhuotao Tian, Xiaojuan Qi, Bei Yu, Hanwang Zhang, Richang Hong
arXiv:2607. 23346v1 Announce Type: new Abstract: Modern deep neural networks are potent catalysts for scientific and industrial impact, yet excessive parameter counts impede deployment in low-compute settings such as hospital equipment and energy infrastructure.
By Aditya Dewan, Arjun Yogeswaran, Benjamin Fedoruk
arXiv:2510. 06048v4 Announce Type: replace Abstract: Effective data selection is essential for pretraining large language models (LLMs), enhancing efficiency and improving generalization to downstream tasks.
By Jie Hao, Rui Yu, Wei Zhang, Huixia Wang, Jie Xu, Mingrui Liu
The paper investigates how the sampled-token reverse-KL loss in on‑policy distillation distributes updates across tokens. By analyzing the gradient of the per‑token K2 estimator, the authors find that tokens with low student probability and large teacher‑student gaps receive disproportionately large gradient norms. They propose Surprise‑aware Reweighting (SuRe), a lightweight weighting rule that further amplifies this allocation, and demonstrate that SuRe improves math metrics on Qwen3 student models without harming out‑of‑domain performance.
By Bing Shao, Jiazheng Zhang, Long Ma, Yujiong Shen, Senjie Jin, Xin Guo, Yuming Yang, Mingxu Chai, Zhiheng Xi, Tao Gui, Qi Zhang, Xuanjing Huang
The paper introduces Stochastic Truncated On-Policy Distillation (STOPD), a method that speeds up on-policy distillation by randomly truncating student rollouts and correcting sampling bias with blockwise inverse-probability weighting. STOPD retains teacher feedback for every generated token and preserves the expected gradient of vanilla OPD while providing theoretical error bounds for common optimizers. Experiments show that STOPD cuts generated tokens by 25–48% and achieves 1.3–2.2× faster training without sacrificing accuracy.
By Zhengxin Zhang, Maosen Tang, Ning Wang, Changhao Li, Jingren Zhou, Bo Dai, Sainyam Galhotra, Alex Townsend, Claire Cardie
arXiv:2607. 22258v1 Announce Type: new Abstract: Deep learning models using traditional softmax classifiers have achieved remarkable success in various classification tasks.
By Yi-Hang Zhu, Rajeev Raman, Shiqi Su, Jianyuan Sun, Xinyu Yang, Nan Xing, Huiyu Zhou
arXiv:2601. 07155v3 Announce Type: replace-cross Abstract: Knowledge distillation (KD) is a widely adopted technique for transferring knowledge from large language models to smaller student models; however, conventional supervised KD often suffers from a distribution mismatch between training and inference.
By Ijun Jang, Jewon Yeom, Juan Yeo, Hyunggyu Lim, Taesup Kim