A Dense Representation Framework for Lexical and Semantic Matching

نویسندگان

چکیده

Lexical and semantic matching capture different successful approaches to text retrieval the fusion of their results has proven be more effective robust than either alone. Prior work performs hybrid by conducting lexical using systems (e.g., Lucene Faiss, respectively) then fusing model outputs. In contrast, our integrates representations with dense densifying high-dimensional into what we call low-dimensional (DLRs). Our experiments show that DLRs can effectively approximate original representations, preserving effectiveness while improving query latency. Furthermore, combine generate (DHRs) are flexible yield faster compared existing techniques. addition, explore jointly training in a single empirically resulting DHRs able advantages individual components. best DHR is competitive state-of-the-art single-vector multi-vector retrievers both in-domain zero-shot evaluation settings. requires smaller indexes, making representation framework an attractive approach retrieval. code available at https://github.com/castorini/dhr .

برای دانلود باید عضویت طلایی داشته باشید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

A Representation Framework for Cross-lingual/Interlingual Lexical Semantic Correspondences

This paper proposes a framework for representing cross-lingual/interlingual lexical semantic correspondences that are expected to be recovered through a series of on-demand/on-the-fly invocations of a lexical semantic matching process. One of the central notions of the proposed framework is a pseudo synset, which is introduced to represent a cross-lingual/multilingual lexical concept, jointly d...

متن کامل

A framework for lexical representation

In this paper we present a unification-based lexical platform designed for highly inflected languages (like Roman ones). A formalism is proposed for encoding a lemma-based lexical source, well suited for linguistic generalizations. From this source, we automatically generate an allomorph indexed dictionary, adequate for efficient processing. A set of software tools have been implemented around ...

متن کامل

Lexical Reference: a Semantic Matching Subtask

Semantic lexical matching is a prominent subtask within text understanding applications. Yet, it is rarely evaluated in a direct manner. This paper proposes a definition for lexical reference which captures the common goals of lexical matching. Based on this definition we created and analyzed a test dataset that was utilized to directly evaluate, compare and improve lexical matching models. We ...

متن کامل

a framework for identifying and prioritizing factors affecting customers’ online shopping behavior in iran

the purpose of this study is identifying effective factors which make customers shop online in iran and investigating the importance of discovered factors in online customers’ decision. in the identifying phase, to discover the factors affecting online shopping behavior of customers in iran, the derived reference model summarizing antecedents of online shopping proposed by change et al. was us...

15 صفحه اول

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: ACM Transactions on Information Systems

سال: 2023

ISSN: ['1558-1152', '1558-2868', '1046-8188', '0734-2047']

DOI: https://doi.org/10.1145/3582426