@inproceedings{gao-etal-2023-precise,
title = "Precise Zero-Shot Dense Retrieval without Relevance Labels",
author = "Gao, Luyu and
Ma, Xueguang and
Lin, Jimmy and
Callan, Jamie",
editor = "Rogers, Anna and
Boyd-Graber, Jordan and
Okazaki, Naoaki",
booktitle = "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.acl-long.99",
doi = "10.18653/v1/2023.acl-long.99",
pages = "1762--1777",
abstract = "While dense retrieval has been shown to be effective and efficient across tasks and languages, it remains difficult to create effective fully zero-shot dense retrieval systems when no relevance labels are available. In this paper, we recognize the difficulty of zero-shot learning and encoding relevance. Instead, we propose to pivot through Hypothetical Document Embeddings (HyDE). Given a query, HyDE first zero-shot prompts an instruction-following language model (e.g., InstructGPT) to generate a hypothetical document. The document captures relevance patterns but is {``}fake{''} and may contain hallucinations. Then, an unsupervised contrastively learned encoder (e.g., Contriever) encodes the document into an embedding vector. This vector identifies a neighborhood in the corpus embedding space, from which similar real documents are retrieved based on vector similarity. This second step grounds the generated document to the actual corpus, with the encoder{'}s dense bottleneck filtering out the hallucinations. Our experiments show that HyDE significantly outperforms the state-of-the-art unsupervised dense retriever Contriever and shows strong performance comparable to fine-tuned retrievers across various tasks (e.g. web search, QA, fact verification) and in non-English languages (e.g., sw, ko, ja, bn).",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="gao-etal-2023-precise">
<titleInfo>
<title>Precise Zero-Shot Dense Retrieval without Relevance Labels</title>
</titleInfo>
<name type="personal">
<namePart type="given">Luyu</namePart>
<namePart type="family">Gao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xueguang</namePart>
<namePart type="family">Ma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jimmy</namePart>
<namePart type="family">Lin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jamie</namePart>
<namePart type="family">Callan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Anna</namePart>
<namePart type="family">Rogers</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jordan</namePart>
<namePart type="family">Boyd-Graber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Naoaki</namePart>
<namePart type="family">Okazaki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Toronto, Canada</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>While dense retrieval has been shown to be effective and efficient across tasks and languages, it remains difficult to create effective fully zero-shot dense retrieval systems when no relevance labels are available. In this paper, we recognize the difficulty of zero-shot learning and encoding relevance. Instead, we propose to pivot through Hypothetical Document Embeddings (HyDE). Given a query, HyDE first zero-shot prompts an instruction-following language model (e.g., InstructGPT) to generate a hypothetical document. The document captures relevance patterns but is “fake” and may contain hallucinations. Then, an unsupervised contrastively learned encoder (e.g., Contriever) encodes the document into an embedding vector. This vector identifies a neighborhood in the corpus embedding space, from which similar real documents are retrieved based on vector similarity. This second step grounds the generated document to the actual corpus, with the encoder’s dense bottleneck filtering out the hallucinations. Our experiments show that HyDE significantly outperforms the state-of-the-art unsupervised dense retriever Contriever and shows strong performance comparable to fine-tuned retrievers across various tasks (e.g. web search, QA, fact verification) and in non-English languages (e.g., sw, ko, ja, bn).</abstract>
<identifier type="citekey">gao-etal-2023-precise</identifier>
<identifier type="doi">10.18653/v1/2023.acl-long.99</identifier>
<location>
<url>https://aclanthology.org/2023.acl-long.99</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>1762</start>
<end>1777</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Precise Zero-Shot Dense Retrieval without Relevance Labels
%A Gao, Luyu
%A Ma, Xueguang
%A Lin, Jimmy
%A Callan, Jamie
%Y Rogers, Anna
%Y Boyd-Graber, Jordan
%Y Okazaki, Naoaki
%S Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F gao-etal-2023-precise
%X While dense retrieval has been shown to be effective and efficient across tasks and languages, it remains difficult to create effective fully zero-shot dense retrieval systems when no relevance labels are available. In this paper, we recognize the difficulty of zero-shot learning and encoding relevance. Instead, we propose to pivot through Hypothetical Document Embeddings (HyDE). Given a query, HyDE first zero-shot prompts an instruction-following language model (e.g., InstructGPT) to generate a hypothetical document. The document captures relevance patterns but is “fake” and may contain hallucinations. Then, an unsupervised contrastively learned encoder (e.g., Contriever) encodes the document into an embedding vector. This vector identifies a neighborhood in the corpus embedding space, from which similar real documents are retrieved based on vector similarity. This second step grounds the generated document to the actual corpus, with the encoder’s dense bottleneck filtering out the hallucinations. Our experiments show that HyDE significantly outperforms the state-of-the-art unsupervised dense retriever Contriever and shows strong performance comparable to fine-tuned retrievers across various tasks (e.g. web search, QA, fact verification) and in non-English languages (e.g., sw, ko, ja, bn).
%R 10.18653/v1/2023.acl-long.99
%U https://aclanthology.org/2023.acl-long.99
%U https://doi.org/10.18653/v1/2023.acl-long.99
%P 1762-1777
Markdown (Informal)
[Precise Zero-Shot Dense Retrieval without Relevance Labels](https://aclanthology.org/2023.acl-long.99) (Gao et al., ACL 2023)
ACL
- Luyu Gao, Xueguang Ma, Jimmy Lin, and Jamie Callan. 2023. Precise Zero-Shot Dense Retrieval without Relevance Labels. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1762–1777, Toronto, Canada. Association for Computational Linguistics.