<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ma-etal-2026-every">
    <titleInfo>
        <title>Every Time I Hire a Linguist, Inference Costs Go Down: On Linguistic Rules as Effective Prompt Compressors</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Jianfei</namePart>
        <namePart type="family">Ma</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Zhaoxin</namePart>
        <namePart type="family">Feng</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Emmanuele</namePart>
        <namePart type="family">Chersoni</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Si</namePart>
        <namePart type="family">Chen</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 2026 Conference on Empirical Methods in Natural Language Processing</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Hung-Yi</namePart>
            <namePart type="family">Lee</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Sunipa</namePart>
            <namePart type="family">Dev</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jessy</namePart>
            <namePart type="family">Li</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Desmond</namePart>
            <namePart type="family">Elliott</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">979-8-89176-459-0</identifier>
    </relatedItem>
    <abstract>Prompt compression shortens LLM input to reduce inference cost, yet existing methods score token importance through LM forward passes. It remains questionable whether such nuanced, costly token selection is necessary. Compression requires identifying informative content, a problem that linguistic research has long addressed through cues that can be operationalized as deterministic rules. We therefore ask: can linguistic rules alone serve as effective prompt compressors, without LM-based scoring at compression time?\par To address this, we conduct offline evolutionary search over lexical, syntactic, semantic, and discourse seeds to find competitive rule combinations. The resulting linguistic compressor requires no LM forward pass at deployment and uses only CPU-side processing for compression. We evaluate it with a dual-path protocol to balance compression quality and reconstruction fidelity.\par Across short passages, multi-document reasoning, and dialogue-memory QA datasets, evolved compressors achieve performance similar to that of recent advanced prompt-compression strategies. Performance is strongest under light-to-moderate compression and degrades as compression becomes more aggressive, while the Direct and Reconstruction paths exhibit distinct patterns. Evolutionary analysis reveals that effective compression fuses signals across linguistic levels and, as the compression ratio increases, rules shift from token pruning to sentence extraction.</abstract>
    <identifier type="citekey">ma-etal-2026-every</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.511/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>11714</start>
            <end>11750</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Every Time I Hire a Linguist, Inference Costs Go Down: On Linguistic Rules as Effective Prompt Compressors
%A Ma, Jianfei
%A Feng, Zhaoxin
%A Chersoni, Emmanuele
%A Chen, Si
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F ma-etal-2026-every
%X Prompt compression shortens LLM input to reduce inference cost, yet existing methods score token importance through LM forward passes. It remains questionable whether such nuanced, costly token selection is necessary. Compression requires identifying informative content, a problem that linguistic research has long addressed through cues that can be operationalized as deterministic rules. We therefore ask: can linguistic rules alone serve as effective prompt compressors, without LM-based scoring at compression time?\par To address this, we conduct offline evolutionary search over lexical, syntactic, semantic, and discourse seeds to find competitive rule combinations. The resulting linguistic compressor requires no LM forward pass at deployment and uses only CPU-side processing for compression. We evaluate it with a dual-path protocol to balance compression quality and reconstruction fidelity.\par Across short passages, multi-document reasoning, and dialogue-memory QA datasets, evolved compressors achieve performance similar to that of recent advanced prompt-compression strategies. Performance is strongest under light-to-moderate compression and degrades as compression becomes more aggressive, while the Direct and Reconstruction paths exhibit distinct patterns. Evolutionary analysis reveals that effective compression fuses signals across linguistic levels and, as the compression ratio increases, rules shift from token pruning to sentence extraction.
%U https://aclanthology.org/2026.emnlp-main.511/
%P 11714-11750

Markdown (Informal)

[Every Time I Hire a Linguist, Inference Costs Go Down: On Linguistic Rules as Effective Prompt Compressors](https://aclanthology.org/2026.emnlp-main.511/) (Ma et al., EMNLP 2026)

• Every Time I Hire a Linguist, Inference Costs Go Down: On Linguistic Rules as Effective Prompt Compressors (Ma et al., EMNLP 2026)

ACL

• Jianfei Ma, Zhaoxin Feng, Emmanuele Chersoni, and Si Chen. 2026. Every Time I Hire a Linguist, Inference Costs Go Down: On Linguistic Rules as Effective Prompt Compressors. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 11714–11750, Budapest, Hungary. Association for Computational Linguistics.