Improving Table Retrieval with Question Generation from Partial Tables

2025-08-08

AI Summary

To address insufficient semantic alignment and low efficiency in full-table embedding for open-domain table retrieval, this paper proposes QGpT: a method that leverages large language models to automatically generate semantically relevant questions from local table fragments (e.g., row/column subsets), and jointly embeds fragments and generated questions into aligned dense representations. QGpT is the first approach to introduce local-structure-driven question generation into table representation learning—bypassing full-table embedding to achieve lightweight yet highly effective query-table semantic matching. Integrated within a late-interaction retrieval framework, QGpT consistently improves recall across multiple benchmark datasets, particularly under complex natural language queries. Moreover, it is fully compatible with mainstream dense retrieval architectures.

Technology Category

• Natural Language Processing: Question Answering
• Search and Optimization: Learning to Search
• Data Mining & Knowledge Management: Intelligent Query Processing

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Recent advances in open-domain question answering over tables have widely adopted large language models (LLMs) under the Retriever-Reader architecture. Prior works have effectively leveraged LLMs to tackle the complex reasoning demands of the Reader component, such as text-to-text, text-to-SQL, and multi hop reasoning. In contrast, the Retriever component has primarily focused on optimizing the query representation-training retrievers to retrieve relevant tables based on questions, or to select keywords from questions for matching table segments. However, little attention has been given to enhancing how tables themselves are represented in embedding space to better align with questions. To address this, we propose QGpT (Question Generation from Partial Tables), a simple yet effective method that uses an LLM to generate synthetic questions based on small portions of a table. These questions are generated to simulate how a user might query the content of the table currently under consideration. The generated questions are then jointly embedded with the partial table segments used for generation, enhancing semantic alignment with user queries. Without the need to embed entire tables, our method significantly improves retrieval performance across multiple benchmarks for both dense and late-interaction retrievers.

Problem

• Enhancing table representation in embedding space
• Improving semantic alignment between tables and user queries
• Generating synthetic questions from partial tables for better retrieval

Innovation

• Generates synthetic questions from partial tables
• Enhances semantic alignment with user queries
• Improves retrieval without embedding entire tables