Logical Consistency is Vital: Neural-Symbolic Information Retrieval for Negative-Constraint Queries

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xu, Ganlin, Zhang, Zhoujia, Mei, Wangyi, Liang, Jiaqing, Lu, Weijia, Zhang, Xiaodong, Yang, Zhifei, Ma, Xiaofeng, Xiao, Yanghua, Yang, Deqing
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910973172908032
author Xu, Ganlin
Zhang, Zhoujia
Mei, Wangyi
Liang, Jiaqing
Lu, Weijia
Zhang, Xiaodong
Yang, Zhifei
Ma, Xiaofeng
Xiao, Yanghua
Yang, Deqing
author_facet Xu, Ganlin
Zhang, Zhoujia
Mei, Wangyi
Liang, Jiaqing
Lu, Weijia
Zhang, Xiaodong
Yang, Zhifei
Ma, Xiaofeng
Xiao, Yanghua
Yang, Deqing
contents Information retrieval plays a crucial role in resource localization. Current dense retrievers retrieve the relevant documents within a corpus via embedding similarities, which compute similarities between dense vectors mainly depending on word co-occurrence between queries and documents, but overlook the real query intents. Thus, they often retrieve numerous irrelevant documents. Particularly in the scenarios of complex queries such as \emph{negative-constraint queries}, their retrieval performance could be catastrophic. To address the issue, we propose a neuro-symbolic information retrieval method, namely \textbf{NS-IR}, that leverages first-order logic (FOL) to optimize the embeddings of naive natural language by considering the \emph{logical consistency} between queries and documents. Specifically, we introduce two novel techniques, \emph{logic alignment} and \emph{connective constraint}, to rerank candidate documents, thereby enhancing retrieval relevance. Furthermore, we construct a new dataset \textbf{NegConstraint} including negative-constraint queries to evaluate our NS-IR's performance on such complex IR scenarios. Our extensive experiments demonstrate that NS-IR not only achieves superior zero-shot retrieval performance on web search and low-resource retrieval tasks, but also performs better on negative-constraint queries. Our scource code and dataset are available at https://github.com/xgl-git/NS-IR-main.
format Preprint
id arxiv_https___arxiv_org_abs_2505_22299
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Logical Consistency is Vital: Neural-Symbolic Information Retrieval for Negative-Constraint Queries
Xu, Ganlin
Zhang, Zhoujia
Mei, Wangyi
Liang, Jiaqing
Lu, Weijia
Zhang, Xiaodong
Yang, Zhifei
Ma, Xiaofeng
Xiao, Yanghua
Yang, Deqing
Information Retrieval
Information retrieval plays a crucial role in resource localization. Current dense retrievers retrieve the relevant documents within a corpus via embedding similarities, which compute similarities between dense vectors mainly depending on word co-occurrence between queries and documents, but overlook the real query intents. Thus, they often retrieve numerous irrelevant documents. Particularly in the scenarios of complex queries such as \emph{negative-constraint queries}, their retrieval performance could be catastrophic. To address the issue, we propose a neuro-symbolic information retrieval method, namely \textbf{NS-IR}, that leverages first-order logic (FOL) to optimize the embeddings of naive natural language by considering the \emph{logical consistency} between queries and documents. Specifically, we introduce two novel techniques, \emph{logic alignment} and \emph{connective constraint}, to rerank candidate documents, thereby enhancing retrieval relevance. Furthermore, we construct a new dataset \textbf{NegConstraint} including negative-constraint queries to evaluate our NS-IR's performance on such complex IR scenarios. Our extensive experiments demonstrate that NS-IR not only achieves superior zero-shot retrieval performance on web search and low-resource retrieval tasks, but also performs better on negative-constraint queries. Our scource code and dataset are available at https://github.com/xgl-git/NS-IR-main.
title Logical Consistency is Vital: Neural-Symbolic Information Retrieval for Negative-Constraint Queries
topic Information Retrieval
url https://arxiv.org/abs/2505.22299