$\texttt{MixGR}$: Enhancing Retriever Generalization for Scientific Domain through Complementary Granularity

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Cai, Fengyu, Zhao, Xinran, Chen, Tong, Chen, Sihao, Zhang, Hongming, Gurevych, Iryna, Koeppl, Heinz
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866915001354158080
author Cai, Fengyu
Zhao, Xinran
Chen, Tong
Chen, Sihao
Zhang, Hongming
Gurevych, Iryna
Koeppl, Heinz
author_facet Cai, Fengyu
Zhao, Xinran
Chen, Tong
Chen, Sihao
Zhang, Hongming
Gurevych, Iryna
Koeppl, Heinz
contents Recent studies show the growing significance of document retrieval in the generation of LLMs, i.e., RAG, within the scientific domain by bridging their knowledge gap. However, dense retrievers often struggle with domain-specific retrieval and complex query-document relationships, particularly when query segments correspond to various parts of a document. To alleviate such prevalent challenges, this paper introduces $\texttt{MixGR}$, which improves dense retrievers' awareness of query-document matching across various levels of granularity in queries and documents using a zero-shot approach. $\texttt{MixGR}$ fuses various metrics based on these granularities to a united score that reflects a comprehensive query-document similarity. Our experiments demonstrate that $\texttt{MixGR}$ outperforms previous document retrieval by 24.7%, 9.8%, and 6.9% on nDCG@5 with unsupervised, supervised, and LLM-based retrievers, respectively, averaged on queries containing multiple subqueries from five scientific retrieval datasets. Moreover, the efficacy of two downstream scientific question-answering tasks highlights the advantage of $\texttt{MixGR}$ to boost the application of LLMs in the scientific domain. The code and experimental datasets are available.
format Preprint
id arxiv_https___arxiv_org_abs_2407_10691
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle $\texttt{MixGR}$: Enhancing Retriever Generalization for Scientific Domain through Complementary Granularity
Cai, Fengyu
Zhao, Xinran
Chen, Tong
Chen, Sihao
Zhang, Hongming
Gurevych, Iryna
Koeppl, Heinz
Information Retrieval
Computation and Language
Recent studies show the growing significance of document retrieval in the generation of LLMs, i.e., RAG, within the scientific domain by bridging their knowledge gap. However, dense retrievers often struggle with domain-specific retrieval and complex query-document relationships, particularly when query segments correspond to various parts of a document. To alleviate such prevalent challenges, this paper introduces $\texttt{MixGR}$, which improves dense retrievers' awareness of query-document matching across various levels of granularity in queries and documents using a zero-shot approach. $\texttt{MixGR}$ fuses various metrics based on these granularities to a united score that reflects a comprehensive query-document similarity. Our experiments demonstrate that $\texttt{MixGR}$ outperforms previous document retrieval by 24.7%, 9.8%, and 6.9% on nDCG@5 with unsupervised, supervised, and LLM-based retrievers, respectively, averaged on queries containing multiple subqueries from five scientific retrieval datasets. Moreover, the efficacy of two downstream scientific question-answering tasks highlights the advantage of $\texttt{MixGR}$ to boost the application of LLMs in the scientific domain. The code and experimental datasets are available.
title $\texttt{MixGR}$: Enhancing Retriever Generalization for Scientific Domain through Complementary Granularity
topic Information Retrieval
Computation and Language
url https://arxiv.org/abs/2407.10691