@inproceedings{zhang-etal-2026-find,
title = "Find Your Optimal Teacher: Personalized Data Synthesis via Router-Guided Multi-Teacher Distillation",
author = "Zhang, Hengyuan and
Yang, Shiping and
Liang, Xiao and
Shang, Chenming and
Jiang, Yuxuan and
Tao, Chaofan and
Xiong, Jing and
So, Hayden Kwok-Hay and
Xie, Ruobing and
Chang, Angel X and
Wong, Ngai",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.acl-long.666/",
pages = "14619--14637",
ISBN = "979-8-89176-390-6",
abstract = "Training student models on synthetic data generated by strong teacher models is a promising approach to distilling the capabilities of teachers. However, existing studies reveal that stronger models are not always optimal teachers, suggesting a mismatch between the teacher{'}s output and the student{'}s learning ability. To address this issue, we propose PerSyn (Personalized data Synthesis), a novel and efficient approach that customizes synthetic data to align with the learning capabilities of the student model. Specifically, our PerSyn method routes each prompt to its optimal teacher via a query-level router that jointly considers the student models' learnability and teacher models' response quality. It successfully transfers the synthesis paradigm from the conventional ``Generate then Select'' to a more efficient manner, i.e., ``Route then Generate'', eliminating the need for all teacher models to generate parallel responses across the entire prompt set. Extensive experiments across different model families and scales demonstrate that PerSyn consistently outperforms all baselines on six benchmarks, including instruct tuning and math reasoning settings. Further analysis verifies the effectiveness of PerSyn and offers extra insights to propel future research. Our code is available at https://anonymous.4open.science/r/PerSyn-8D85."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhang-etal-2026-find">
<titleInfo>
<title>Find Your Optimal Teacher: Personalized Data Synthesis via Router-Guided Multi-Teacher Distillation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hengyuan</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shiping</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiao</namePart>
<namePart type="family">Liang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chenming</namePart>
<namePart type="family">Shang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuxuan</namePart>
<namePart type="family">Jiang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chaofan</namePart>
<namePart type="family">Tao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jing</namePart>
<namePart type="family">Xiong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hayden</namePart>
<namePart type="given">Kwok-Hay</namePart>
<namePart type="family">So</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruobing</namePart>
<namePart type="family">Xie</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Angel</namePart>
<namePart type="given">X</namePart>
<namePart type="family">Chang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ngai</namePart>
<namePart type="family">Wong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Liakata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viviane</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Moreira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Jurgens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-390-6</identifier>
</relatedItem>
<abstract>Training student models on synthetic data generated by strong teacher models is a promising approach to distilling the capabilities of teachers. However, existing studies reveal that stronger models are not always optimal teachers, suggesting a mismatch between the teacher’s output and the student’s learning ability. To address this issue, we propose PerSyn (Personalized data Synthesis), a novel and efficient approach that customizes synthetic data to align with the learning capabilities of the student model. Specifically, our PerSyn method routes each prompt to its optimal teacher via a query-level router that jointly considers the student models’ learnability and teacher models’ response quality. It successfully transfers the synthesis paradigm from the conventional “Generate then Select” to a more efficient manner, i.e., “Route then Generate”, eliminating the need for all teacher models to generate parallel responses across the entire prompt set. Extensive experiments across different model families and scales demonstrate that PerSyn consistently outperforms all baselines on six benchmarks, including instruct tuning and math reasoning settings. Further analysis verifies the effectiveness of PerSyn and offers extra insights to propel future research. Our code is available at https://anonymous.4open.science/r/PerSyn-8D85.</abstract>
<identifier type="citekey">zhang-etal-2026-find</identifier>
<location>
<url>https://aclanthology.org/2026.acl-long.666/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>14619</start>
<end>14637</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Find Your Optimal Teacher: Personalized Data Synthesis via Router-Guided Multi-Teacher Distillation
%A Zhang, Hengyuan
%A Yang, Shiping
%A Liang, Xiao
%A Shang, Chenming
%A Jiang, Yuxuan
%A Tao, Chaofan
%A Xiong, Jing
%A So, Hayden Kwok-Hay
%A Xie, Ruobing
%A Chang, Angel X.
%A Wong, Ngai
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-390-6
%F zhang-etal-2026-find
%X Training student models on synthetic data generated by strong teacher models is a promising approach to distilling the capabilities of teachers. However, existing studies reveal that stronger models are not always optimal teachers, suggesting a mismatch between the teacher’s output and the student’s learning ability. To address this issue, we propose PerSyn (Personalized data Synthesis), a novel and efficient approach that customizes synthetic data to align with the learning capabilities of the student model. Specifically, our PerSyn method routes each prompt to its optimal teacher via a query-level router that jointly considers the student models’ learnability and teacher models’ response quality. It successfully transfers the synthesis paradigm from the conventional “Generate then Select” to a more efficient manner, i.e., “Route then Generate”, eliminating the need for all teacher models to generate parallel responses across the entire prompt set. Extensive experiments across different model families and scales demonstrate that PerSyn consistently outperforms all baselines on six benchmarks, including instruct tuning and math reasoning settings. Further analysis verifies the effectiveness of PerSyn and offers extra insights to propel future research. Our code is available at https://anonymous.4open.science/r/PerSyn-8D85.
%U https://aclanthology.org/2026.acl-long.666/
%P 14619-14637
Markdown (Informal)
[Find Your Optimal Teacher: Personalized Data Synthesis via Router-Guided Multi-Teacher Distillation](https://aclanthology.org/2026.acl-long.666/) (Zhang et al., ACL 2026)
ACL
- Hengyuan Zhang, Shiping Yang, Xiao Liang, Chenming Shang, Yuxuan Jiang, Chaofan Tao, Jing Xiong, Hayden Kwok-Hay So, Ruobing Xie, Angel X Chang, and Ngai Wong. 2026. Find Your Optimal Teacher: Personalized Data Synthesis via Router-Guided Multi-Teacher Distillation. In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 14619–14637, San Diego, California, United States. Association for Computational Linguistics.