Enregistré dans:
Détails bibliographiques
Auteurs principaux: Zhang, Hao, Zhang, Zhibin, Wu, Guangxin, Ning, Wanyi, Guo, Jiafeng, Cheng, Xueqi
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:https://arxiv.org/abs/2605.01732
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866914527165022208
author Zhang, Hao
Zhang, Zhibin
Wu, Guangxin
Ning, Wanyi
Guo, Jiafeng
Cheng, Xueqi
author_facet Zhang, Hao
Zhang, Zhibin
Wu, Guangxin
Ning, Wanyi
Guo, Jiafeng
Cheng, Xueqi
contents 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 distillation offers a promising solution by transferring knowledge from a large teacher model to a smaller student model. However, existing distillation methods typically treat all tokens equally, ignoring the fact that different tokens contribute unequally to model decisions. This can lead to inefficient knowledge transfer and reduced learning effectiveness. To address this limitation, we propose an entropy-based adaptive distillation strategy that dynamically adjusts the training process at the token level. Our method leverages the teacher's output entropy to guide three aspects of distillation. Specifically, we introduce a token-level curriculum by dynamically shifting focus from low- to high-entropy tokens during training. We further adjust the distillation temperature based on token entropy to better capture teacher confidence patterns. Moreover, we employ a dual-branch architecture for efficient logits-only distillation on easy tokens and deeper feature-based distillation on difficult tokens. Extensive experiments validate the soundness and effectiveness of our method.
format Preprint
id arxiv_https___arxiv_org_abs_2605_01732
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle EGAD: Entropy-Guided Adaptive Distillation for Token-Level Knowledge Transfer
Zhang, Hao
Zhang, Zhibin
Wu, Guangxin
Ning, Wanyi
Guo, Jiafeng
Cheng, Xueqi
Computation and Language
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 distillation offers a promising solution by transferring knowledge from a large teacher model to a smaller student model. However, existing distillation methods typically treat all tokens equally, ignoring the fact that different tokens contribute unequally to model decisions. This can lead to inefficient knowledge transfer and reduced learning effectiveness. To address this limitation, we propose an entropy-based adaptive distillation strategy that dynamically adjusts the training process at the token level. Our method leverages the teacher's output entropy to guide three aspects of distillation. Specifically, we introduce a token-level curriculum by dynamically shifting focus from low- to high-entropy tokens during training. We further adjust the distillation temperature based on token entropy to better capture teacher confidence patterns. Moreover, we employ a dual-branch architecture for efficient logits-only distillation on easy tokens and deeper feature-based distillation on difficult tokens. Extensive experiments validate the soundness and effectiveness of our method.
title EGAD: Entropy-Guided Adaptive Distillation for Token-Level Knowledge Transfer
topic Computation and Language
url https://arxiv.org/abs/2605.01732