@inproceedings{wang-etal-2026-grasprune,
title = "{GRASP}rune: Global Gating for Budgeted Structured Pruning of Large Language Models",
author = "Wang, Ziyang and
Xiao, Jiangfeng and
Xiao, Chuan and
LI, Ruoxiang and
Mao, Rui and
Qin, Jianbin",
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.491/",
pages = "10719--10736",
ISBN = "979-8-89176-390-6",
abstract = "Large language models (LLMs) are expensive to serve because dense FFN blocks, multi-head attention, and KV caches dominate memory, making structured pruning a natural way to reduce serving costs under tight parameter and memory budgets. We present GRASPrune, a global budgeted structured pruning framework applied post-hoc to a pretrained model that jointly prunes FFN channels and attention KV head groups under a single global parameter budget. GRASPrune attaches lightweight learnable gates to prunable units and optimizes only these gates on a small unlabeled language-modeling calibration set, keeping all backbone weights frozen while enforcing the target sparsity at every step. A final budget-preserving scaling calibration reweights the surviving channels and heads to correct scale shifts introduced by pruning. On LLaMA-2-7B, GRASPrune removes 50{\%} of parameters and achieves 12.18 perplexity on WikiText-2 while maintaining competitive average zero-shot accuracy on five downstream benchmarks, using a short calibration run of four epochs on 512 unlabeled sequences on a single NVIDIA A100 80GB GPU, all without any full-model fine-tuning."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2026-grasprune">
<titleInfo>
<title>GRASPrune: Global Gating for Budgeted Structured Pruning of Large Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ziyang</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiangfeng</namePart>
<namePart type="family">Xiao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chuan</namePart>
<namePart type="family">Xiao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruoxiang</namePart>
<namePart type="family">LI</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rui</namePart>
<namePart type="family">Mao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jianbin</namePart>
<namePart type="family">Qin</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>Large language models (LLMs) are expensive to serve because dense FFN blocks, multi-head attention, and KV caches dominate memory, making structured pruning a natural way to reduce serving costs under tight parameter and memory budgets. We present GRASPrune, a global budgeted structured pruning framework applied post-hoc to a pretrained model that jointly prunes FFN channels and attention KV head groups under a single global parameter budget. GRASPrune attaches lightweight learnable gates to prunable units and optimizes only these gates on a small unlabeled language-modeling calibration set, keeping all backbone weights frozen while enforcing the target sparsity at every step. A final budget-preserving scaling calibration reweights the surviving channels and heads to correct scale shifts introduced by pruning. On LLaMA-2-7B, GRASPrune removes 50% of parameters and achieves 12.18 perplexity on WikiText-2 while maintaining competitive average zero-shot accuracy on five downstream benchmarks, using a short calibration run of four epochs on 512 unlabeled sequences on a single NVIDIA A100 80GB GPU, all without any full-model fine-tuning.</abstract>
<identifier type="citekey">wang-etal-2026-grasprune</identifier>
<location>
<url>https://aclanthology.org/2026.acl-long.491/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>10719</start>
<end>10736</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T GRASPrune: Global Gating for Budgeted Structured Pruning of Large Language Models
%A Wang, Ziyang
%A Xiao, Jiangfeng
%A Xiao, Chuan
%A LI, Ruoxiang
%A Mao, Rui
%A Qin, Jianbin
%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 wang-etal-2026-grasprune
%X Large language models (LLMs) are expensive to serve because dense FFN blocks, multi-head attention, and KV caches dominate memory, making structured pruning a natural way to reduce serving costs under tight parameter and memory budgets. We present GRASPrune, a global budgeted structured pruning framework applied post-hoc to a pretrained model that jointly prunes FFN channels and attention KV head groups under a single global parameter budget. GRASPrune attaches lightweight learnable gates to prunable units and optimizes only these gates on a small unlabeled language-modeling calibration set, keeping all backbone weights frozen while enforcing the target sparsity at every step. A final budget-preserving scaling calibration reweights the surviving channels and heads to correct scale shifts introduced by pruning. On LLaMA-2-7B, GRASPrune removes 50% of parameters and achieves 12.18 perplexity on WikiText-2 while maintaining competitive average zero-shot accuracy on five downstream benchmarks, using a short calibration run of four epochs on 512 unlabeled sequences on a single NVIDIA A100 80GB GPU, all without any full-model fine-tuning.
%U https://aclanthology.org/2026.acl-long.491/
%P 10719-10736
Markdown (Informal)
[GRASPrune: Global Gating for Budgeted Structured Pruning of Large Language Models](https://aclanthology.org/2026.acl-long.491/) (Wang et al., ACL 2026)
ACL