English
Related papers

Related papers: Understanding and Improving Knowledge Distillation…

200 papers

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

Recently, knowledge distillation (KD) has shown great success in BERT compression. Instead of only learning from the teacher's soft label as in conventional KD, researchers find that the rich information contained in the hidden layers of…

Computation and Language · Computer Science 2021-06-11 Yuanxin Liu , Fandong Meng , Zheng Lin , Weiping Wang , Jie Zhou

Knowledge Distillation (KD) utilizes training data as a transfer set to transfer knowledge from a complex network (Teacher) to a smaller network (Student). Several works have recently identified many scenarios where the training data may…

Computer Vision and Pattern Recognition · Computer Science 2021-10-28 Gaurav Kumar Nayak , Monish Keswani , Sharan Seshadri , Anirban Chakraborty

Benefiting from well-trained deep neural networks (DNNs), model compression have captured special attention for computing resource limited equipment, especially edge devices. Knowledge distillation (KD) is one of the widely used compression…

Machine Learning · Computer Science 2024-06-06 Jinyin Chen , Xiaoming Zhao , Haibin Zheng , Xiao Li , Sheng Xiang , Haifeng Guo

With the emergence of Mixture-of-Experts (MoE), the efficient scaling of model size has accelerated the development of large language models in recent years. However, their high memory requirements prevent their use in resource-constrained…

Computation and Language · Computer Science 2025-02-19 Gyeongman Kim , Gyouk Chu , Eunho Yang

Knowledge distillation aims to compress a powerful yet cumbersome teacher model into a lightweight student model without much sacrifice of performance. For this purpose, various approaches have been proposed over the past few years,…

Computer Vision and Pattern Recognition · Computer Science 2022-03-29 Defang Chen , Jian-Ping Mei , Hailin Zhang , Can Wang , Yan Feng , Chun Chen

Speech denoising is a generally adopted and impactful task, appearing in many common and everyday-life use cases. Although there are very powerful methods published, most of those are too complex for deployment in everyday and low-resources…

Sound · Computer Science 2025-05-07 Diep Luong , Mikko Heikkinen , Konstantinos Drossos , Tuomas Virtanen

Knowledge distillation (KD) is known as a promising solution to compress large language models (LLMs) via transferring their knowledge to smaller models. During this process, white-box KD methods usually minimize the distance between the…

Computation and Language · Computer Science 2024-10-02 Songming Zhang , Xue Zhang , Zengkui Sun , Yufeng Chen , Jinan Xu

Recent studies pointed out that knowledge distillation (KD) suffers from two degradation problems, the teacher-student gap and the incompatibility with strong data augmentations, making it not applicable to training state-of-the-art models,…

Machine Learning · Computer Science 2022-02-17 Jihao Liu , Boxiao Liu , Hongsheng Li , Yu Liu

Although large foundation models pre-trained by self-supervised learning have achieved state-of-the-art performance in many tasks including automatic speech recognition (ASR), knowledge distillation (KD) is often required in practice to…

Audio and Speech Processing · Electrical Eng. & Systems 2023-03-21 Xiaoyu Yang , Qiujia Li , Chao Zhang , Philip C. Woodland

Typical technique in knowledge distillation (KD) is regularizing the learning of a limited capacity model (student) by pushing its responses to match a powerful model's (teacher). Albeit useful especially in the penultimate layer and…

Computer Vision and Pattern Recognition · Computer Science 2023-09-08 Ada Gorgun , Yeti Z. Gurbuz , A. Aydin Alatan

Beyond the complexity of CNNs that require training on large annotated datasets, the domain shift between design and operational data has limited the adoption of CNNs in many real-world applications. For instance, in person…

Computer Vision and Pattern Recognition · Computer Science 2021-01-20 Le Thanh Nguyen-Meidine , Atif Belal , Madhu Kiran , Jose Dolz , Louis-Antoine Blais-Morin , Eric Granger

Knowledge distillation (KD) is widely used for compressing a teacher model to reduce its inference cost and memory footprint, by training a smaller student model. However, current KD methods for auto-regressive sequence models suffer from…

Machine Learning · Computer Science 2024-01-18 Rishabh Agarwal , Nino Vieillard , Yongchao Zhou , Piotr Stanczyk , Sabela Ramos , Matthieu Geist , Olivier Bachem

Knowledge distillation is to transfer the knowledge from the data learned by the teacher network to the student network, so that the student has the advantage of less parameters and less calculations, and the accuracy is close to the…

Machine Learning · Computer Science 2020-06-03 Zaida Zhou , Chaoran Zhuge , Xinwei Guan , Wen Liu

Knowledge distillation (KD) has been widely used to transfer knowledge from large, accurate models (teachers) to smaller, efficient ones (students). Recent methods have explored enforcing consistency by incorporating causal interpretations…

Computer Vision and Pattern Recognition · Computer Science 2025-07-17 Nikolaos Giakoumoglou , Tania Stathaki

Pre-trained Transformer models such as BERT have shown great success in a wide range of applications, but at the cost of substantial increases in model complexity. Quantization-aware training (QAT) is a promising method to lower the…

Computation and Language · Computer Science 2023-02-24 Minsoo Kim , Kyuhong Shim , Seongmin Park , Wonyong Sung , Jungwook Choi

Knowledge Distillation (KD) consists of transferring “knowledge” from one machine learning model (the teacher) to another (the student). Commonly, the teacher is a high-capacity model with formidable performance, while the student is…

Machine Learning · Statistics 2024-03-05 Tommaso Furlanello , Zachary C. Lipton , Michael Tschannen , Laurent Itti , Anima Anandkumar

Fine-tuned transformer models have shown superior performances in many natural language tasks. However, the large model size prohibits deploying high-performance transformer models on resource-constrained devices. This paper proposes a…

Computation and Language · Computer Science 2024-10-01 Zi Yang , Samridhi Choudhary , Siegfried Kunzmann , Zheng Zhang

We investigate the effects of post-training quantization and quantization-aware training on the generalization of Transformer language models. We present a new method called self-distilled quantization (SDQ) that minimizes accumulative…

Computation and Language · Computer Science 2023-07-13 James O' Neill , Sourav Dutta

Knowledge distillation (KD) is a technique used to transfer knowledge from an overparameterized teacher network to a less-parameterized student network, thereby minimizing the incurred performance loss. KD methods can be categorized into…

Computer Vision and Pattern Recognition · Computer Science 2024-06-17 Jaeyeon Jang , Young-Ik Kim , Jisu Lim , Hyeonseong Lee