<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="fakhrutdinov-2026-pare4bit">
    <titleInfo>
        <title>Pare4Bit: Quantization, Depth Pruning, and Model Selection for the WMT26 Model Compression Task</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Nail</namePart>
        <namePart type="family">Fakhrutdinov</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-10</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the Eleventh Conference on Machine Translation</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Barry</namePart>
            <namePart type="family">Haddow</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Tom</namePart>
            <namePart type="family">Kocmi</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Philipp</namePart>
            <namePart type="family">Koehn</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Christof</namePart>
            <namePart type="family">Monz</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Budapest, Hungary</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-474-3</identifier>
    </relatedItem>
    <abstract>In the constrained track we compress google/gemma-3-12b-it with a data-light pipeline—vision-tower removal followed by GPTQ W4A16 quantization - reaching 43.2 mean chrF at 7.1 GB, a 3.4× size reduction for a 0.3 chrF drop. We further explore depth pruning with LoRA knowledge-distillation healing, which recovers a collapsed pruned model to near-parity and adds an engine-agnostic speedup. In the unconstrained track we treat model selection as a first-class compression decision: the newer Gemma-4-12B, deployed via quantization-aware-trained INT4, outperforms the constrained Gemma-3 baseline by +2.5 mean chrF.</abstract>
    <identifier type="citekey">fakhrutdinov-2026-pare4bit</identifier>
    <location>
        <url>https://aclanthology.org/2026.wmt-main.135/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>1986</start>
            <end>1990</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Pare4Bit: Quantization, Depth Pruning, and Model Selection for the WMT26 Model Compression Task
%A Fakhrutdinov, Nail
%Y Haddow, Barry
%Y Kocmi, Tom
%Y Koehn, Philipp
%Y Monz, Christof
%S Proceedings of the Eleventh Conference on Machine Translation
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-474-3
%F fakhrutdinov-2026-pare4bit
%X In the constrained track we compress google/gemma-3-12b-it with a data-light pipeline—vision-tower removal followed by GPTQ W4A16 quantization - reaching 43.2 mean chrF at 7.1 GB, a 3.4× size reduction for a 0.3 chrF drop. We further explore depth pruning with LoRA knowledge-distillation healing, which recovers a collapsed pruned model to near-parity and adds an engine-agnostic speedup. In the unconstrained track we treat model selection as a first-class compression decision: the newer Gemma-4-12B, deployed via quantization-aware-trained INT4, outperforms the constrained Gemma-3 baseline by +2.5 mean chrF.
%U https://aclanthology.org/2026.wmt-main.135/
%P 1986-1990

Markdown (Informal)

[Pare4Bit: Quantization, Depth Pruning, and Model Selection for the WMT26 Model Compression Task](https://aclanthology.org/2026.wmt-main.135/) (Fakhrutdinov, WMT 2026)

• Pare4Bit: Quantization, Depth Pruning, and Model Selection for the WMT26 Model Compression Task (Fakhrutdinov, WMT 2026)

ACL

• Nail Fakhrutdinov. 2026. Pare4Bit: Quantization, Depth Pruning, and Model Selection for the WMT26 Model Compression Task. In Proceedings of the Eleventh Conference on Machine Translation, pages 1986–1990, Budapest, Hungary. Association for Computational Linguistics.