English
Related papers

Related papers: LLM-Oriented Token-Adaptive Knowledge Distillation

200 papers

This paper studies the problem of pre-training for small models, which is essential for many mobile devices. Current state-of-the-art methods on this problem transfer the representational knowledge of a large network (as a Teacher) into a…

Computer Vision and Pattern Recognition · Computer Science 2024-06-18 Mingsheng Li , Lin Zhang , Mingzhen Zhu , Zilong Huang , Gang Yu , Jiayuan Fan , Tao Chen

This work studies knowledge distillation (KD) for large language models (LLMs) through preference optimization. We propose a reward-guided imitation learning framework for sequential KD, formulating a min-max optimization problem between…

Machine Learning · Computer Science 2025-05-27 Chen Jia

Growing efforts to improve knowledge distillation (KD) in large language models (LLMs) replace dense teacher supervision with selective distillation, which uses a subset of token positions, vocabulary classes, or training samples for…

Computation and Language · Computer Science 2026-02-03 Almog Tavor , Itay Ebenspanger , Neil Cnaan , Mor Geva

Knowledge Distillation (KD) has emerged as a promising approach for transferring knowledge from a larger, more complex teacher model to a smaller student model. Traditionally, KD involves training the student to mimic the teacher's output…

Machine Learning · Computer Science 2024-10-03 Noel Loo , Fotis Iliopoulos , Wei Hu , Erik Vee

Data-free knowledge distillation (DFKD) is a widely-used strategy for Knowledge Distillation (KD) whose training data is not available. It trains a lightweight student model with the aid of a large pretrained teacher model without any…

Computer Vision and Pattern Recognition · Computer Science 2023-07-06 Jingru Li , Sheng Zhou , Liangcheng Li , Haishuai Wang , Zhi Yu , Jiajun Bu

Discrete diffusion models (DDMs) have shown powerful generation ability for discrete data modalities like text and molecules. However, their practical application is hindered by inefficient sampling, requiring a large number of sampling…

Machine Learning · Computer Science 2025-09-25 Feiyang Fu , Tongxian Guo , Zhaoqiang Liu

Knowledge Distillation (KD) is a critical tool for training Large Language Models (LLMs), yet the majority of research focuses on approaches that rely solely on output logits, neglecting semantic information in the teacher's intermediate…

Computation and Language · Computer Science 2026-05-13 Maxime Guigon , Lucas Dixon , Michaël E. Sander

Online Knowledge Distillation (OKD) methods streamline the distillation training process into a single stage, eliminating the need for knowledge transfer from a pretrained teacher network to a more compact student network. This paper…

Computer Vision and Pattern Recognition · Computer Science 2025-05-06 Zhaowei Chen , Borui Zhao , Yuchen Ge , Yuhao Chen , Renjie Song , Jiajun Liang

We propose Cross-Attention-based Non-local Knowledge Distillation (CanKD), a novel feature-based knowledge distillation framework that leverages cross-attention mechanisms to enhance the knowledge transfer process. Unlike traditional…

Computer Vision and Pattern Recognition · Computer Science 2025-11-27 Shizhe Sun , Wataru Ohyama

With numerous medical tasks, the performance of deep models has recently experienced considerable improvements. These models are often adept learners. Yet, their intricate architectural design and high computational complexity make…

Image and Video Processing · Electrical Eng. & Systems 2023-03-17 Eddardaa Ben Loussaief , Hatem Rashwan , Mohammed Ayad , Mohammed Zakaria Hassan , Domenec Puig

Recent research on knowledge distillation has increasingly focused on logit distillation because of its simplicity, effectiveness, and versatility in model compression. In this paper, we introduce Refined Logit Distillation (RLD) to address…

Computer Vision and Pattern Recognition · Computer Science 2025-07-29 Wujie Sun , Defang Chen , Siwei Lyu , Genlang Chen , Chun Chen , Can Wang

Knowledge distillation (KD) is a valuable yet challenging approach that enhances a compact student network by learning from a high-performance but cumbersome teacher model. However, previous KD methods for image restoration overlook the…

Computer Vision and Pattern Recognition · Computer Science 2024-12-18 Yunshuai Zhou , Junbo Qiao , Jincheng Liao , Wei Li , Simiao Li , Jiao Xie , Yunhang Shen , Jie Hu , Shaohui Lin

Although large models have shown a strong capacity to solve large-scale problems in many areas including natural language and computer vision, their voluminous parameters are hard to deploy in a real-time system due to computational and…

Machine Learning · Computer Science 2025-01-07 Sirong Wu , Xi Luo , Junjie Liu , Yuhui Deng

Knowledge distillation is a method of transferring the knowledge from a complex deep neural network (DNN) to a smaller and faster DNN, while preserving its accuracy. Recent variants of knowledge distillation include teaching assistant…

Machine Learning · Computer Science 2023-04-11 Minghong Gao

Knowledge Distillation (KD) aims at transferring the knowledge of a well-performed neural network (the {\it teacher}) to a weaker one (the {\it student}). A peculiar phenomenon is that a more accurate model doesn't necessarily teach better,…

Machine Learning · Computer Science 2022-10-14 Xin-Chun Li , Wen-Shu Fan , Shaoming Song , Yinchuan Li , Bingshuai Li , Yunfeng Shao , De-Chuan Zhan

Large Vision-Language Models (VLMs) are successful in addressing a multitude of vision-language understanding tasks, such as Visual Question Answering (VQA), but their memory and compute requirements remain a concern for practical…

Computer Vision and Pattern Recognition · Computer Science 2026-05-12 Nikolaos Gkalelis , Vasileios Mezaris

To reduce a model size but retain performance, we often rely on knowledge distillation (KD) which transfers knowledge from a large "teacher" model to a smaller "student" model. However, KD on multimodal datasets such as vision-language…

Computer Vision and Pattern Recognition · Computer Science 2021-10-25 Woojeong Jin , Maziar Sanjabi , Shaoliang Nie , Liang Tan , Xiang Ren , Hamed Firooz

Knowledge distillation (KD) has become a prevalent technique for compressing large language models (LLMs). Existing KD methods are constrained by the need for identical tokenizers (i.e., vocabularies) between teacher and student models,…

Computation and Language · Computer Science 2025-01-22 Xiao Cui , Mo Zhu , Yulei Qin , Liang Xie , Wengang Zhou , Houqiang Li

Knowledge Distillation (KD) aims to learn a compact student network using knowledge from a large pre-trained teacher network, where both networks are trained on data from the same distribution. However, in practical applications, the…

Machine Learning · Computer Science 2024-01-17 Jialiang Tang , Shuo Chen , Gang Niu , Hongyuan Zhu , Joey Tianyi Zhou , Chen Gong , Masashi Sugiyama

Pre-trained language models have been applied to various NLP tasks with considerable performance gains. However, the large model sizes, together with the long inference time, limit the deployment of such models in real-time applications.…

Computation and Language · Computer Science 2022-11-03 Haojie Pan , Chengyu Wang , Minghui Qiu , Yichang Zhang , Yaliang Li , Jun Huang