English
Related papers

Related papers: On-Policy Context Distillation for Language Models

200 papers

Knowledge Distillation (KD) has been considered as a key solution in model compression and acceleration in recent years. In KD, a small student model is generally trained from a large teacher model by minimizing the divergence between the…

Machine Learning · Computer Science 2021-11-16 Raed Alharbi , Minh N. Vu , My T. Thai

Small language models (SLMs) are crucial for applications with strict latency and computational constraints, yet achieving high performance remains challenging. Knowledge distillation (KD) can transfer capabilities from large teacher…

Computation and Language · Computer Science 2026-03-20 Jingyu Peng , Maolin Wang , Hengyi Cai , Yuchen Li , Kai Zhang , Shuaiqiang Wang , Dawei Yin , Xiangyu Zhao

Knowledge distillation is a widely used technique for model compression. We posit that the teacher model used in a distillation setup, captures relationships between classes, that extend beyond the original dataset. We empirically show that…

Machine Learning · Computer Science 2019-11-14 Gaurav Menghani , Sujith Ravi

Recent advancements in large language models (LLMs) have raised concerns about inference costs, increasing the need for research into model compression. While knowledge distillation (KD) is a prominent method for this, research on KD for…

Computation and Language · Computer Science 2024-09-30 Gyeongman Kim , Doohyuk Jang , Eunho Yang

Knowledge distillation aims to transfer representation ability from a teacher model to a student model. Previous approaches focus on either individual representation distillation or inter-sample similarity preservation. While we argue that…

Computer Vision and Pattern Recognition · Computer Science 2021-03-31 Jinguo Zhu , Shixiang Tang , Dapeng Chen , Shijie Yu , Yakun Liu , Aijun Yang , Mingzhe Rong , Xiaohua Wang

Deep learning achieved great progress recently, however, it is not easy or efficient to further improve its performance by increasing the size of the model. Multi-modal learning can mitigate this challenge by introducing richer and more…

Artificial Intelligence · Computer Science 2025-10-07 Cairong Zhao , Yufeng Jin , Zifan Song , Haonan Chen , Duoqian Miao , Guosheng Hu

The burgeoning complexity of contemporary deep learning models, while achieving unparalleled accuracy, has inadvertently introduced deployment challenges in resource-constrained environments. Knowledge distillation, a technique aiming to…

Machine Learning · Computer Science 2023-10-05 Sia Gholami , Marwan Omar

Knowledge distillation typically involves transferring knowledge from a Large Language Model (LLM) to a Smaller Language Model (SLM). However, in tasks such as text matching, fine-tuned smaller models often yield more effective…

Computation and Language · Computer Science 2025-07-09 Mingzhe Li , Jing Xiang , Qishen Zhang , Kaiyang Wan , Xiuying Chen

The emergence of large-scale pre-trained models has heightened their application in various downstream tasks, yet deployment is a challenge in environments with limited computational resources. Knowledge distillation has emerged as a…

Machine Learning · Computer Science 2024-06-27 Jinbin Huang , Wenbin He , Liang Gou , Liu Ren , Chris Bryan

On-policy self-distillation trains a reasoning model on its own rollouts while a teacher, often the same model conditioned on privileged context, provides dense token-level supervision. Existing objectives typically weight the teacher's…

Artificial Intelligence · Computer Science 2026-05-14 Junlong Ke , Zichen Wen , Weijia Li , Conghui He , Linfeng Zhang

Large-scale language models have recently demonstrated impressive empirical performance. Nevertheless, the improved results are attained at the price of bigger models, more power consumption, and slower inference, which hinder their…

Computation and Language · Computer Science 2021-03-18 Kevin J Liang , Weituo Hao , Dinghan Shen , Yufan Zhou , Weizhu Chen , Changyou Chen , Lawrence Carin

Modern language models have the capacity to store and use immense amounts of knowledge about real-world entities, but it remains unclear how to update such knowledge stored in model parameters. While prior methods for updating knowledge in…

Computation and Language · Computer Science 2023-11-01 Shankar Padmanabhan , Yasumasa Onoe , Michael J. Q. Zhang , Greg Durrett , Eunsol Choi

In natural language processing (NLP) tasks, slow inference speed and huge footprints in GPU usage remain the bottleneck of applying pre-trained deep models in production. As a popular method for model compression, knowledge distillation…

Computation and Language · Computer Science 2020-12-15 Fei Yuan , Linjun Shou , Jian Pei , Wutao Lin , Ming Gong , Yan Fu , Daxin Jiang

Significant advancements in image generation have been made with diffusion models. Nevertheless, when contrasted with previous generative models, diffusion models face substantial computational overhead, leading to failure in real-time…

Computer Vision and Pattern Recognition · Computer Science 2024-10-08 Linfeng Zhang , Kaisheng Ma

Consistency distillation is a prevalent way for accelerating diffusion models adopted in consistency (trajectory) models, in which a student model is trained to traverse backward on the probability flow (PF) ordinary differential equation…

Machine Learning · Computer Science 2025-05-01 Kaiwen Zheng , Guande He , Jianfei Chen , Fan Bao , Jun Zhu

Online Continual Learning (OCL) addresses the problem of training neural networks on a continuous data stream where multiple classification tasks emerge in sequence. In contrast to offline Continual Learning, data can be seen only once in…

Machine Learning · Computer Science 2024-06-06 Nicolas Michel , Maorong Wang , Ling Xiao , Toshihiko Yamasaki

On-policy self-distillation (OPSD) trains a student on its own rollouts using a privileged teacher, but its standard objective weights all generated tokens equally, implicitly treating the privileged teacher target as equally reliable at…

Machine Learning · Computer Science 2026-05-22 Xiaogeng Liu , Xinyan Wang , Yingzi Ma , Yechao Zhang , Chaowei Xiao

Knowledge distillation (KD) is a standard route to compress Large Language Models (LLMs) into compact students, yet most pipelines uniformly apply token-wise loss regardless of teacher confidence. This indiscriminate supervision amplifies…

Computation and Language · Computer Science 2025-11-18 Haiduo Huang , Jiangcheng Song , Yadong Zhang , Pengju Ren

On-policy knowledge distillation (OPD) trains a student on its own rollouts under token-level supervision from a teacher. Not all token positions matter equally, but existing views of token importance are incomplete. We ask a direct…

Machine Learning · Computer Science 2026-05-22 Yuanda Xu , Hejian Sang , Zhengze Zhou , Ran He , Zhipeng Wang , Alborz Geramifard

Practical autonomous driving systems face two crucial challenges: memory constraints and domain gap issues. In this paper, we present a novel approach to learn domain adaptive knowledge in models with limited memory, thus bestowing the…

Computer Vision and Pattern Recognition · Computer Science 2020-11-30 Divya Kothandaraman , Athira Nambiar , Anurag Mittal