<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lin-2025-operational">
    <titleInfo>
        <title>Operational Advice for Dense and Sparse Retrievers: HNSW, Flat, or Inverted Indexes?</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Jimmy</namePart>
        <namePart type="family">Lin</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025-07</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 6: Industry Track)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Georg</namePart>
            <namePart type="family">Rehm</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Yunyao</namePart>
            <namePart type="family">Li</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Vienna, Austria</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-288-6</identifier>
    </relatedItem>
    <abstract>Practitioners working on dense retrieval today face a bewildering number of choices. Beyond selecting the embedding model, another consequential choice is the actual implementation of nearest-neighbor vector search. While best practices recommend HNSW indexes, flat vector indexes with brute-force search represent another viable option, particularly for smaller corpora and for rapid prototyping. In this paper, we provide experimental results on the BEIR dataset using the open-source Lucene search library that explicate the tradeoffs between HNSW and flat indexes (including quantized variants) from the perspectives of indexing time, query evaluation performance, and retrieval quality. With additional comparisons between dense and sparse retrievers, our results provide guidance for today’s search practitioner in understanding the design space of dense and sparse retrievers. To our knowledge, we are the first to provide operational advice supported by empirical experiments in this regard.</abstract>
    <identifier type="citekey">lin-2025-operational</identifier>
    <identifier type="doi">10.18653/v1/2025.acl-industry.61</identifier>
    <location>
        <url>https://aclanthology.org/2025.acl-industry.61/</url>
    </location>
    <part>
        <date>2025-07</date>
        <extent unit="page">
            <start>865</start>
            <end>872</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Operational Advice for Dense and Sparse Retrievers: HNSW, Flat, or Inverted Indexes?
%A Lin, Jimmy
%Y Rehm, Georg
%Y Li, Yunyao
%S Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 6: Industry Track)
%D 2025
%8 July
%I Association for Computational Linguistics
%C Vienna, Austria
%@ 979-8-89176-288-6
%F lin-2025-operational
%X Practitioners working on dense retrieval today face a bewildering number of choices. Beyond selecting the embedding model, another consequential choice is the actual implementation of nearest-neighbor vector search. While best practices recommend HNSW indexes, flat vector indexes with brute-force search represent another viable option, particularly for smaller corpora and for rapid prototyping. In this paper, we provide experimental results on the BEIR dataset using the open-source Lucene search library that explicate the tradeoffs between HNSW and flat indexes (including quantized variants) from the perspectives of indexing time, query evaluation performance, and retrieval quality. With additional comparisons between dense and sparse retrievers, our results provide guidance for today’s search practitioner in understanding the design space of dense and sparse retrievers. To our knowledge, we are the first to provide operational advice supported by empirical experiments in this regard.
%R 10.18653/v1/2025.acl-industry.61
%U https://aclanthology.org/2025.acl-industry.61/
%U https://doi.org/10.18653/v1/2025.acl-industry.61
%P 865-872

Markdown (Informal)

[Operational Advice for Dense and Sparse Retrievers: HNSW, Flat, or Inverted Indexes?](https://aclanthology.org/2025.acl-industry.61/) (Lin, ACL 2025)

• Operational Advice for Dense and Sparse Retrievers: HNSW, Flat, or Inverted Indexes? (Lin, ACL 2025)

ACL

• Jimmy Lin. 2025. Operational Advice for Dense and Sparse Retrievers: HNSW, Flat, or Inverted Indexes?. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 6: Industry Track), pages 865–872, Vienna, Austria. Association for Computational Linguistics.