@inproceedings{salim-etal-2026-expert,
title = "Expert Calibration Lens for Pruning Mixture of Experts",
author = "Salim, Luis Frentzen and
Wu, Chia-Chun and
Van Nhiem, Tran and
Ku, Lun-Wei and
Li, Yung-Hui",
editor = "Durrett, Greg and
Jian, Ping",
booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 3: System Demonstrations)",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.acl-demo.72/",
pages = "736--742",
ISBN = "979-8-89176-392-0",
abstract = "Expert pruning is a practical deployment technique for Mixture-of-Experts (MoE) models. It reduces resource usage and mitigates expert redundancy, but its success depends strongly on the calibration set used for pruning. In domain-general settings, it is unclear which properties of the calibration data drive good pruning outcomes, and the effects of calibration perturbations are often unintuitive. We observe, for example, that calibration sets in different languages can lead to very similar pruning results despite appearing dissimilar on the surface.To address this, we propose Expert Calibration Lens, a lightweight analysis tool that compares expert activation patterns across datasets to predict the impact of calibration perturbations without repeatedly running expensive pruning procedures. We use activations that are quick to compute and evaluate the resulting analysis for downstream task performance."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="salim-etal-2026-expert">
<titleInfo>
<title>Expert Calibration Lens for Pruning Mixture of Experts</title>
</titleInfo>
<name type="personal">
<namePart type="given">Luis</namePart>
<namePart type="given">Frentzen</namePart>
<namePart type="family">Salim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chia-Chun</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tran</namePart>
<namePart type="family">Van Nhiem</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lun-Wei</namePart>
<namePart type="family">Ku</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yung-Hui</namePart>
<namePart type="family">Li</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 3: System Demonstrations)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Greg</namePart>
<namePart type="family">Durrett</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ping</namePart>
<namePart type="family">Jian</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">979-8-89176-392-0</identifier>
</relatedItem>
<abstract>Expert pruning is a practical deployment technique for Mixture-of-Experts (MoE) models. It reduces resource usage and mitigates expert redundancy, but its success depends strongly on the calibration set used for pruning. In domain-general settings, it is unclear which properties of the calibration data drive good pruning outcomes, and the effects of calibration perturbations are often unintuitive. We observe, for example, that calibration sets in different languages can lead to very similar pruning results despite appearing dissimilar on the surface.To address this, we propose Expert Calibration Lens, a lightweight analysis tool that compares expert activation patterns across datasets to predict the impact of calibration perturbations without repeatedly running expensive pruning procedures. We use activations that are quick to compute and evaluate the resulting analysis for downstream task performance.</abstract>
<identifier type="citekey">salim-etal-2026-expert</identifier>
<location>
<url>https://aclanthology.org/2026.acl-demo.72/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>736</start>
<end>742</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Expert Calibration Lens for Pruning Mixture of Experts
%A Salim, Luis Frentzen
%A Wu, Chia-Chun
%A Van Nhiem, Tran
%A Ku, Lun-Wei
%A Li, Yung-Hui
%Y Durrett, Greg
%Y Jian, Ping
%S Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 3: System Demonstrations)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-392-0
%F salim-etal-2026-expert
%X Expert pruning is a practical deployment technique for Mixture-of-Experts (MoE) models. It reduces resource usage and mitigates expert redundancy, but its success depends strongly on the calibration set used for pruning. In domain-general settings, it is unclear which properties of the calibration data drive good pruning outcomes, and the effects of calibration perturbations are often unintuitive. We observe, for example, that calibration sets in different languages can lead to very similar pruning results despite appearing dissimilar on the surface.To address this, we propose Expert Calibration Lens, a lightweight analysis tool that compares expert activation patterns across datasets to predict the impact of calibration perturbations without repeatedly running expensive pruning procedures. We use activations that are quick to compute and evaluate the resulting analysis for downstream task performance.
%U https://aclanthology.org/2026.acl-demo.72/
%P 736-742
Markdown (Informal)
[Expert Calibration Lens for Pruning Mixture of Experts](https://aclanthology.org/2026.acl-demo.72/) (Salim et al., ACL 2026)
ACL
- Luis Frentzen Salim, Chia-Chun Wu, Tran Van Nhiem, Lun-Wei Ku, and Yung-Hui Li. 2026. Expert Calibration Lens for Pruning Mixture of Experts. In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 3: System Demonstrations), pages 736–742, San Diego, California, United States. Association for Computational Linguistics.