<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhao-etal-2025-fistapruner">
    <titleInfo>
        <title>FISTAPruner: Layer-wise Post-training Pruning for Large Language Models</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Pengxiang</namePart>
        <namePart type="family">Zhao</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Hanyu</namePart>
        <namePart type="family">Hu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Ping</namePart>
        <namePart type="family">Li</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yi</namePart>
        <namePart type="family">Zheng</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Zhefeng</namePart>
        <namePart type="family">Wang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Xiaoming</namePart>
        <namePart type="family">Yuan</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>Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing</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">979-8-89176-332-6</identifier>
    </relatedItem>
    <abstract>Pruning is a critical strategy for compressing trained large language models (LLMs), aiming at substantial memory conservation and computational acceleration without compromising performance. However, existing pruning methods typically necessitate inefficient retraining for billion-scale LLMs or rely on heuristically designed metrics to determine pruning masks, leading to performance degradation. This paper presents, for the first time, a LASSO-like convex optimization model crafted to induce sparsity in LLMs. By leveraging FISTA, we introduce FISTAPruner, a novel method that includes a cumulative error elimination mechanism within decoder layers and supports parallel pruning for unstructured pruning. Additionally, we extend this method to 2:4 semi-structured pruning. We comprehensively evaluate FISTAPruner on models such as OPT, LLaMA, and Qwen variants with 125M to 70B parameters under unstructured and 2:4 semi-structured sparsity, showcasing superior performance over existing methods across various language benchmarks. Notably, it can remove 50% of the model parameters for LLaMA-3-70B while retaining 98.6% and 95.6% of the zero-shot task performance under these two sparsity patterns, respectively.</abstract>
    <identifier type="citekey">zhao-etal-2025-fistapruner</identifier>
    <identifier type="doi">10.18653/v1/2025.emnlp-main.1498</identifier>
    <location>
        <url>https://aclanthology.org/2025.emnlp-main.1498/</url>
    </location>
    <part>
        <date>2025-11</date>
        <extent unit="page">
            <start>29469</start>
            <end>29487</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T FISTAPruner: Layer-wise Post-training Pruning for Large Language Models
%A Zhao, Pengxiang
%A Hu, Hanyu
%A Li, Ping
%A Zheng, Yi
%A Wang, Zhefeng
%A Yuan, Xiaoming
%Y Christodoulopoulos, Christos
%Y Chakraborty, Tanmoy
%Y Rose, Carolyn
%Y Peng, Violet
%S Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-332-6
%F zhao-etal-2025-fistapruner
%X Pruning is a critical strategy for compressing trained large language models (LLMs), aiming at substantial memory conservation and computational acceleration without compromising performance. However, existing pruning methods typically necessitate inefficient retraining for billion-scale LLMs or rely on heuristically designed metrics to determine pruning masks, leading to performance degradation. This paper presents, for the first time, a LASSO-like convex optimization model crafted to induce sparsity in LLMs. By leveraging FISTA, we introduce FISTAPruner, a novel method that includes a cumulative error elimination mechanism within decoder layers and supports parallel pruning for unstructured pruning. Additionally, we extend this method to 2:4 semi-structured pruning. We comprehensively evaluate FISTAPruner on models such as OPT, LLaMA, and Qwen variants with 125M to 70B parameters under unstructured and 2:4 semi-structured sparsity, showcasing superior performance over existing methods across various language benchmarks. Notably, it can remove 50% of the model parameters for LLaMA-3-70B while retaining 98.6% and 95.6% of the zero-shot task performance under these two sparsity patterns, respectively.
%R 10.18653/v1/2025.emnlp-main.1498
%U https://aclanthology.org/2025.emnlp-main.1498/
%U https://doi.org/10.18653/v1/2025.emnlp-main.1498
%P 29469-29487

Markdown (Informal)

[FISTAPruner: Layer-wise Post-training Pruning for Large Language Models](https://aclanthology.org/2025.emnlp-main.1498/) (Zhao et al., EMNLP 2025)

• FISTAPruner: Layer-wise Post-training Pruning for Large Language Models (Zhao et al., EMNLP 2025)

ACL

• Pengxiang Zhao, Hanyu Hu, Ping Li, Yi Zheng, Zhefeng Wang, and Xiaoming Yuan. 2025. FISTAPruner: Layer-wise Post-training Pruning for Large Language Models. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 29469–29487, Suzhou, China. Association for Computational Linguistics.