@inproceedings{zhang-liu-2025-staged,
title = "Staged Knowledge Distillation Through Least-to-Most Prompting: Optimizing Teacher Guidance via Difficulty-Aware Training",
author = "Zhang, Mengxiang and
Liu, Lingyuan",
editor = "Christodoulopoulos, Christos and
Chakraborty, Tanmoy and
Rose, Carolyn and
Peng, Violet",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2025",
month = nov,
year = "2025",
address = "Suzhou, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-emnlp.451/",
pages = "8489--8501",
ISBN = "979-8-89176-335-7",
abstract = "Knowledge distillation (KD) enables the compression of large language models (LLMs) by transferring knowledge from a high-capacity teacher model to a resource-efficient student model, maintaining competitive performance for tasks such as instruction following. However, conventional white-box KD methods often suffer from training-inference mismatches and suboptimal performance due to the asymmetric nature of Kullback-Leibler divergence (KLD) and reliance on computationally expensive student-generated outputs. To address these challenges, we propose Least-to-Most Prompting Knowledge Distillation (L2M-KD), a novel white-box KD method grounded in curriculum learning (CL) and adaptive loss design. L2M-KD employs a two-pronged approach: (1) a CL strategy that ranks training samples by difficulty using Rouge-L scores, partitioning them into easy-to-hard subsets across multiple stages, and (2) an adaptive KD loss that transitions from KLD to skew KLD, dynamically adjusting teacher guidance to mitigate mode-averaging and over-smoothing. Extensive experiments on instruction-following tasks demonstrate that L2M-KD outperforms existing white-box KD methods, achieving superior student model performance with reduced computational overhead by leveraging ground-truth outputs exclusively. Our findings underscore the efficacy of difficulty-aware training and adaptive teacher guidance, offering a computationally efficient and robust approach to LLM compression."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhang-liu-2025-staged">
<titleInfo>
<title>Staged Knowledge Distillation Through Least-to-Most Prompting: Optimizing Teacher Guidance via Difficulty-Aware Training</title>
</titleInfo>
<name type="personal">
<namePart type="given">Mengxiang</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lingyuan</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2025</title>
</titleInfo>
<name type="personal">
<namePart type="given">Christos</namePart>
<namePart type="family">Christodoulopoulos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tanmoy</namePart>
<namePart type="family">Chakraborty</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Carolyn</namePart>
<namePart type="family">Rose</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Violet</namePart>
<namePart type="family">Peng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Suzhou, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-335-7</identifier>
</relatedItem>
<abstract>Knowledge distillation (KD) enables the compression of large language models (LLMs) by transferring knowledge from a high-capacity teacher model to a resource-efficient student model, maintaining competitive performance for tasks such as instruction following. However, conventional white-box KD methods often suffer from training-inference mismatches and suboptimal performance due to the asymmetric nature of Kullback-Leibler divergence (KLD) and reliance on computationally expensive student-generated outputs. To address these challenges, we propose Least-to-Most Prompting Knowledge Distillation (L2M-KD), a novel white-box KD method grounded in curriculum learning (CL) and adaptive loss design. L2M-KD employs a two-pronged approach: (1) a CL strategy that ranks training samples by difficulty using Rouge-L scores, partitioning them into easy-to-hard subsets across multiple stages, and (2) an adaptive KD loss that transitions from KLD to skew KLD, dynamically adjusting teacher guidance to mitigate mode-averaging and over-smoothing. Extensive experiments on instruction-following tasks demonstrate that L2M-KD outperforms existing white-box KD methods, achieving superior student model performance with reduced computational overhead by leveraging ground-truth outputs exclusively. Our findings underscore the efficacy of difficulty-aware training and adaptive teacher guidance, offering a computationally efficient and robust approach to LLM compression.</abstract>
<identifier type="citekey">zhang-liu-2025-staged</identifier>
<location>
<url>https://aclanthology.org/2025.findings-emnlp.451/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>8489</start>
<end>8501</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Staged Knowledge Distillation Through Least-to-Most Prompting: Optimizing Teacher Guidance via Difficulty-Aware Training
%A Zhang, Mengxiang
%A Liu, Lingyuan
%Y Christodoulopoulos, Christos
%Y Chakraborty, Tanmoy
%Y Rose, Carolyn
%Y Peng, Violet
%S Findings of the Association for Computational Linguistics: EMNLP 2025
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-335-7
%F zhang-liu-2025-staged
%X Knowledge distillation (KD) enables the compression of large language models (LLMs) by transferring knowledge from a high-capacity teacher model to a resource-efficient student model, maintaining competitive performance for tasks such as instruction following. However, conventional white-box KD methods often suffer from training-inference mismatches and suboptimal performance due to the asymmetric nature of Kullback-Leibler divergence (KLD) and reliance on computationally expensive student-generated outputs. To address these challenges, we propose Least-to-Most Prompting Knowledge Distillation (L2M-KD), a novel white-box KD method grounded in curriculum learning (CL) and adaptive loss design. L2M-KD employs a two-pronged approach: (1) a CL strategy that ranks training samples by difficulty using Rouge-L scores, partitioning them into easy-to-hard subsets across multiple stages, and (2) an adaptive KD loss that transitions from KLD to skew KLD, dynamically adjusting teacher guidance to mitigate mode-averaging and over-smoothing. Extensive experiments on instruction-following tasks demonstrate that L2M-KD outperforms existing white-box KD methods, achieving superior student model performance with reduced computational overhead by leveraging ground-truth outputs exclusively. Our findings underscore the efficacy of difficulty-aware training and adaptive teacher guidance, offering a computationally efficient and robust approach to LLM compression.
%U https://aclanthology.org/2025.findings-emnlp.451/
%P 8489-8501
Markdown (Informal)
[Staged Knowledge Distillation Through Least-to-Most Prompting: Optimizing Teacher Guidance via Difficulty-Aware Training](https://aclanthology.org/2025.findings-emnlp.451/) (Zhang & Liu, Findings 2025)
ACL