paper-with-me

Papers

Curriculum Temperature for Knowledge Distillation

2022-11-29 · Zheng Li, Xiang Li, Lingfeng Yang, Borui Zhao, RenJie Song, Lei Luo, Jun Li, Jian Yang

Most existing distillation methods ignore the flexible role of the temperature in the loss function and fix it as a hyper-parameter that can be decided by an inefficient grid search. In general, the temperature controls the discrepancy between two distributions and can faithfully determine the difficulty level of the distillation task. Keeping a constant temperature, i.e., a fixed level of task difficulty, is usually sub-optimal for a growing student during its progressive learning stages. In this paper, we propose a simple curriculum-based technique, termed Curriculum Temperature for Knowledge Distillation (CTKD), which controls the task difficulty level during the student's learning career through a dynamic and learnable temperature. Specifically, following an easy-to-hard curriculum, we gradually increase the distillation loss w.r.t. the temperature, leading to increased distillation difficulty in an adversarial manner. As an easy-to-use plug-in technique, CTKD can be seamlessly integrated into existing knowledge distillation frameworks and brings general improvements at a negligible additional computation cost. Extensive experiments on CIFAR-100, ImageNet-2012, and MS-COCO demonstrate the effectiveness of our method. Our code is available at https://github.com/zhengli97/CTKD.

📄 PDF Abstract BibTeX arXiv:2211.16231

Code (1)

zhengli97/ctkd 공식 구현 pytorch

Tasks

Image ClassificationKnowledge Distillation

Methods 이 논문이 사용한 방법론

Knowledge Distillation A very simple way to improve the performance of almost any machine learning algorithm is to train many different models on the same data and then to average their predictions.…

Similar Papers 제목 키워드 기반

EGAD: Entropy-Guided Adaptive Distillation for Token-Level Knowledge Transfer

2026-05-03 · Hao Zhang, Zhibin Zhang, Guangxin Wu, Wanyi Ning 외 arxiv

Large language models (LLMs) have achieved remarkable performance across diverse domains, yet their enormous computational and memory requirements hinder deployment in resource-constrained environments. Knowledge distill…

Knowledge Distillation

A Survey on Recent Teacher-student Learning Studies

2023-04-10 · Minghong Gao

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 teac…

Knowledge DistillationSurvey

Combining Curriculum Learning and Knowledge Distillation for Dialogue Generation

2021-11-01 · Findings (EMNLP) 2021 11 · Qingqing Zhu, Xiuying Chen, Pengfei Wu, Junfei Liu 외

Curriculum learning, a machine training strategy that feeds training instances to the model from easy to hard, has been proven to facilitate the dialogue generation task. Meanwhile, knowledge distillation, a knowledge tr…

Dialogue GenerationKnowledge DistillationSentence

Teach Me to Trick: Exploring Adversarial Transferability via Knowledge Distillation

2025-07-29 · Siddhartha Pradhan, Shikshya Shiwakoti, Neha Bathuri arxiv

We investigate whether knowledge distillation (KD) from multiple heterogeneous teacher models can enhance the generation of transferable adversarial examples. A lightweight student model is trained using two KD strategie…

Knowledge DistillationModel Compression

Being Strong Progressively! Enhancing Knowledge Distillation of Large Language Models through a Curriculum Learning Framework

2025-06-06 · Lingyuan Liu, Mengxiang Zhang

Knowledge Distillation (KD) compresses large language models (LLMs) by transferring the teacher model's capabilities to a smaller student model, reducing inference cost and memory usage while maintaining performance. How…

Instruction FollowingKnowledge Distillation