Retrieve Only Relevant Tables Whether Few or Many: Adaptive Table Retrieval Method

TL;DR

This paper introduces an adaptive table retrieval method that dynamically adjusts the number of tables retrieved for each query, improving accuracy over fixed top-k approaches.

Contribution

The proposed method uses adaptive thresholding and reranking to better select relevant tables, addressing limitations of fixed-size retrieval strategies.

Findings

Outperforms existing fixed top-k retrieval methods on multiple datasets.

Improves downstream task performance with adaptive retrieval.

Effectively handles large table corpora with reranking.

Abstract

Retrieving relevant tables from extensive databases for a given natural language query is essential for accurately answering questions in tasks such as text-to-SQL. Existing table retrieval approaches select a pre-determined set of k tables with the highest similarity to the query. However, the number of required tables varies across queries and cannot be known in advance. Enforcing a fixed number of retrieved tables regardless of the query may either retrieve an undersized set, failing to obtain all necessary evidence, or retrieve an oversized pool, including irrelevant tables. To address this issue, we propose an adaptive table retrieval method that adjusts the number of tables retrieved according to the requirements of each query. Specifically, we utilize an adaptive thresholding mechanism to selectively retrieve tables and integrate a sliding-window reranking algorithm to efficiently process a large table corpus. Extensive experiments on Spider, BIRD, and Spider 2.0 demonstrate that our method effectively addresses the limitations of the top-k retrieval strategy, improving performance in retrieval and downstream tasks. Our code and data are available at https://github.com/sbY99/Adaptive-Table-Retrieval.