<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="tyagi-etal-2026-beyond">
    <titleInfo>
        <title>Beyond Compressibility: Transferable Low-Dimensional Structure and the Limits of Compression in Financial Retrieval</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Oddharak</namePart>
        <namePart type="family">Tyagi</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Shashishekar</namePart>
        <namePart type="family">Ramakrishna</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Chetan</namePart>
        <namePart type="family">Harsha</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Sridhar</namePart>
        <namePart type="family">Dasaratha</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 11th Workshop on Financial Technology and Natural Language Processing</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Jimin</namePart>
            <namePart type="family">Huang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Yongjae</namePart>
            <namePart type="family">Lee</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Sophia</namePart>
            <namePart type="family">Ananiadou</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Hsin-Hsi</namePart>
            <namePart type="family">Chen</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Hiroya</namePart>
            <namePart type="family">Takamura</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Chung-Chi</namePart>
            <namePart type="family">Chen</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">979-8-89176-466-8</identifier>
    </relatedItem>
    <abstract>Modern retrieval embeddings are highly compressible, often retaining most of their effectiveness after the removal of a large fraction of dimensions. This suggests that retrieval-relevant information is organized in a lower-dimensional structure, but it remains unclear how broadly that structure generalizes and what ultimately limits compression.\par We study this question in long-document financial retrieval where relevant evidence must be distinguished from many highly similar passages containing repeated entities, financial metrics, tables, and boilerplate. Across three financial retrieval benchmarks and eight retrievers, we find that PCA projections learned on one corpus transfer effectively to others, retaining most of the effectiveness of corpus-specific fitting. Transfer quality is strongly associated with principal-subspace overlap, consistent with substantial shared low-dimensional structure across financial corpora. We further show that global geometric measures such as intrinsic dimension only partially explain retrieval robustness under dimensionality reduction. A case study on a financial question answering benchmark links retrieval failures to dense clusters of competing table chunks: Removing these distractors substantially reduces performance degradation. These results suggest that transferability is enabled by low-dimensional structure that generalizes across financial corpora, while compression failures are associated with local retrieval competition that is not captured by existing global geometric measures.</abstract>
    <identifier type="citekey">tyagi-etal-2026-beyond</identifier>
    <location>
        <url>https://aclanthology.org/2026.finnlp-1.34/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>408</start>
            <end>423</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Beyond Compressibility: Transferable Low-Dimensional Structure and the Limits of Compression in Financial Retrieval
%A Tyagi, Oddharak
%A Ramakrishna, Shashishekar
%A Harsha, Chetan
%A Dasaratha, Sridhar
%Y Huang, Jimin
%Y Lee, Yongjae
%Y Ananiadou, Sophia
%Y Chen, Hsin-Hsi
%Y Takamura, Hiroya
%Y Chen, Chung-Chi
%S Proceedings of the 11th Workshop on Financial Technology and Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-466-8
%F tyagi-etal-2026-beyond
%X Modern retrieval embeddings are highly compressible, often retaining most of their effectiveness after the removal of a large fraction of dimensions. This suggests that retrieval-relevant information is organized in a lower-dimensional structure, but it remains unclear how broadly that structure generalizes and what ultimately limits compression.\par We study this question in long-document financial retrieval where relevant evidence must be distinguished from many highly similar passages containing repeated entities, financial metrics, tables, and boilerplate. Across three financial retrieval benchmarks and eight retrievers, we find that PCA projections learned on one corpus transfer effectively to others, retaining most of the effectiveness of corpus-specific fitting. Transfer quality is strongly associated with principal-subspace overlap, consistent with substantial shared low-dimensional structure across financial corpora. We further show that global geometric measures such as intrinsic dimension only partially explain retrieval robustness under dimensionality reduction. A case study on a financial question answering benchmark links retrieval failures to dense clusters of competing table chunks: Removing these distractors substantially reduces performance degradation. These results suggest that transferability is enabled by low-dimensional structure that generalizes across financial corpora, while compression failures are associated with local retrieval competition that is not captured by existing global geometric measures.
%U https://aclanthology.org/2026.finnlp-1.34/
%P 408-423

Markdown (Informal)

[Beyond Compressibility: Transferable Low-Dimensional Structure and the Limits of Compression in Financial Retrieval](https://aclanthology.org/2026.finnlp-1.34/) (Tyagi et al., FinNLP 2026)

• Beyond Compressibility: Transferable Low-Dimensional Structure and the Limits of Compression in Financial Retrieval (Tyagi et al., FinNLP 2026)

ACL

• Oddharak Tyagi, Shashishekar Ramakrishna, Chetan Harsha, and Sridhar Dasaratha. 2026. Beyond Compressibility: Transferable Low-Dimensional Structure and the Limits of Compression in Financial Retrieval. In Proceedings of the 11th Workshop on Financial Technology and Natural Language Processing, pages 408–423, Budapest, Hungary. Association for Computational Linguistics.