Reasoning-Aware Query-Focused Summarization over Multi-Table Data

TL;DR

The paper tackles the problem of generating concise, query-focused summaries from multi-table data that require cross-table reasoning. It introduces QueryTableSummarizer++, an end-to-end generative framework that integrates table-aware pre-training (including row-column masking and inter-table relationship learning), query-aligned fine-tuning, and reinforcement learning with feedback to optimize summary quality. On a newly constructed multi-table benchmark, the method achieves state-of-the-art BLEU, ROUGE, and F1 scores and is supported by human evaluations. The results demonstrate strong domain generalization and scalability to varying table counts, suggesting practical impact for automated analytics over structured data.

Abstract

Query-focused summarization over multi-table data is a challenging yet critical task for extracting precise and relevant information from structured data. Existing methods often rely on complex preprocessing steps and struggle to generalize across domains or handle the logical reasoning required for multi-table queries. In this paper, we propose QueryTableSummarizer++, an end-to-end generative framework leveraging large language models (LLMs) enhanced with table-aware pre-training, query-aligned fine-tuning, and reinforcement learning with feedback. Our method eliminates the need for intermediate serialization steps and directly generates query-relevant summaries. Experiments on a benchmark dataset demonstrate that QueryTableSummarizer++ significantly outperforms state-of-the-art baselines in terms of BLEU, ROUGE, and F1-score. Additional analyses highlight its scalability, generalization across domains, and robust handling of complex queries. Human evaluation further validates the superior quality and practical applicability of the generated summaries, establishing QueryTableSummarizer++ as a highly effective solution for multi-table summarization tasks.