Saved in:
Bibliographic Details
Main Authors: Zhang, Chen, Wang, Jianghui, Cheng, Bingyang, Chen, Zhongtao, XU, Wendong, Wang, Cong, Canini, Marco, Orabona, Francesco, WU, Yik Chung, Wong, Ngai
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.20461
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908849796022272
author Zhang, Chen
Wang, Jianghui
Cheng, Bingyang
Chen, Zhongtao
XU, Wendong
Wang, Cong
Canini, Marco
Orabona, Francesco
WU, Yik Chung
Wong, Ngai
author_facet Zhang, Chen
Wang, Jianghui
Cheng, Bingyang
Chen, Zhongtao
XU, Wendong
Wang, Cong
Canini, Marco
Orabona, Francesco
WU, Yik Chung
Wong, Ngai
contents Attention learners, neural networks built on the attention mechanism, e.g., transformers, excel at learning the implicit relationships that relate sequences to their corresponding properties, e.g., mapping a given sequence of tokens to the probability of the next token. However, the learning process tends to be costly. To address this, we present a novel paradigm named Attention Neural Teaching (AtteNT) that reinterprets the learning process through a nonparametric teaching perspective. Specifically, the latter provides a theoretical framework for teaching mappings that are implicitly defined (i.e., nonparametric) via example selection. Such an implicit mapping is embodied through a dense set of sequence-property pairs, with the AtteNT teacher selecting a subset to accelerate convergence in attention learner training. By analytically investigating the role of attention on parameter-based gradient descent during training, and recasting the evolution of attention learners, shaped by parameter updates, through functional gradient descent in nonparametric teaching, we show for the first time that teaching attention learners is consistent with teaching importance-adaptive nonparametric learners. These new findings readily commit AtteNT to enhancing learning efficiency of attention learners. Specifically, we observe training time reductions of 13.01% for LLMs and 20.58% for ViTs, spanning both fine-tuning and training-from-scratch regimes. Crucially, these gains are achieved without compromising accuracy; in fact, performance is consistently preserved and often enhanced across a diverse set of downstream tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2602_20461
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Nonparametric Teaching of Attention Learners
Zhang, Chen
Wang, Jianghui
Cheng, Bingyang
Chen, Zhongtao
XU, Wendong
Wang, Cong
Canini, Marco
Orabona, Francesco
WU, Yik Chung
Wong, Ngai
Machine Learning
Attention learners, neural networks built on the attention mechanism, e.g., transformers, excel at learning the implicit relationships that relate sequences to their corresponding properties, e.g., mapping a given sequence of tokens to the probability of the next token. However, the learning process tends to be costly. To address this, we present a novel paradigm named Attention Neural Teaching (AtteNT) that reinterprets the learning process through a nonparametric teaching perspective. Specifically, the latter provides a theoretical framework for teaching mappings that are implicitly defined (i.e., nonparametric) via example selection. Such an implicit mapping is embodied through a dense set of sequence-property pairs, with the AtteNT teacher selecting a subset to accelerate convergence in attention learner training. By analytically investigating the role of attention on parameter-based gradient descent during training, and recasting the evolution of attention learners, shaped by parameter updates, through functional gradient descent in nonparametric teaching, we show for the first time that teaching attention learners is consistent with teaching importance-adaptive nonparametric learners. These new findings readily commit AtteNT to enhancing learning efficiency of attention learners. Specifically, we observe training time reductions of 13.01% for LLMs and 20.58% for ViTs, spanning both fine-tuning and training-from-scratch regimes. Crucially, these gains are achieved without compromising accuracy; in fact, performance is consistently preserved and often enhanced across a diverse set of downstream tasks.
title Nonparametric Teaching of Attention Learners
topic Machine Learning
url https://arxiv.org/abs/2602.20461