<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="palomino-2026-layer">
    <titleInfo>
        <title>Layer-Aware Native Quantization for Compact Machine Translation: A WMT26 English-to-Chinese Submission</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Alonso</namePart>
        <namePart type="family">Palomino</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>This paper describes a submission to the constrained English-to-Simplified-Chinese track of the WMT26 Model Compression shared task. Starting from Gemma~3 12B, the system applies native BitsAndBytes NF4 with double quantization to the 144 gate, up, and down projections in the model’s 48 text-transformer MLP blocks. Attention projections, embeddings, normalization layers, the tied language-model head, and vision modules remain in BF16. The method requires no calibration or fine-tuning data and produces a mixed checkpoint whose quantized weights remain low-bit at inference. On the official 945-segment blind set, the system scores 0.7669 CometKiwi-XXL and 2.7836 MetricX-24-XXL, slightly improving on the organizer global-q4 baseline’s 0.7665 and 2.8147, respectively. Its 11.8~GB artifact is 48.4\textbackslash% of BF16 size, and H100 throughput is essentially tied with global q4 (866.9 versus 867.9 source characters/s). On 332 reference-backed WMT25 segments, its chrF and BLEU gains over global q4 are significant under paired bootstrap resampling. A transferred selected-layer q3+Zstd codec reaches statistically indistinguishable local quality but expands to BF16-like memory after decoding.</abstract>
    <identifier type="citekey">palomino-2026-layer</identifier>
    <location>
        <url>https://aclanthology.org/2026.wmt-main.140/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>2025</start>
            <end>2029</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Layer-Aware Native Quantization for Compact Machine Translation: A WMT26 English-to-Chinese Submission
%A Palomino, Alonso
%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 palomino-2026-layer
%X This paper describes a submission to the constrained English-to-Simplified-Chinese track of the WMT26 Model Compression shared task. Starting from Gemma~3 12B, the system applies native BitsAndBytes NF4 with double quantization to the 144 gate, up, and down projections in the model’s 48 text-transformer MLP blocks. Attention projections, embeddings, normalization layers, the tied language-model head, and vision modules remain in BF16. The method requires no calibration or fine-tuning data and produces a mixed checkpoint whose quantized weights remain low-bit at inference. On the official 945-segment blind set, the system scores 0.7669 CometKiwi-XXL and 2.7836 MetricX-24-XXL, slightly improving on the organizer global-q4 baseline’s 0.7665 and 2.8147, respectively. Its 11.8~GB artifact is 48.4\textbackslash% of BF16 size, and H100 throughput is essentially tied with global q4 (866.9 versus 867.9 source characters/s). On 332 reference-backed WMT25 segments, its chrF and BLEU gains over global q4 are significant under paired bootstrap resampling. A transferred selected-layer q3+Zstd codec reaches statistically indistinguishable local quality but expands to BF16-like memory after decoding.
%U https://aclanthology.org/2026.wmt-main.140/
%P 2025-2029

Markdown (Informal)

[Layer-Aware Native Quantization for Compact Machine Translation: A WMT26 English-to-Chinese Submission](https://aclanthology.org/2026.wmt-main.140/) (Palomino, WMT 2026)

• Layer-Aware Native Quantization for Compact Machine Translation: A WMT26 English-to-Chinese Submission (Palomino, WMT 2026)

ACL

• Alonso Palomino. 2026. Layer-Aware Native Quantization for Compact Machine Translation: A WMT26 English-to-Chinese Submission. In Proceedings of the Eleventh Conference on Machine Translation, pages 2025–2029, Budapest, Hungary. Association for Computational Linguistics.