DETQUS: Decomposition-Enhanced Transformers for QUery-focused Summarization

2025-03-07

AI Summary

To address inaccurate summarization in query-driven table-to-text generation caused by Transformer token limits and high inference complexity on large tables, this paper proposes a table decomposition–enhanced paradigm. First, an LLM performs query-aware column selection to compress the input; second, a table-structure-aware OmniTab QA model guides fine-grained preservation of critical information. The method employs a fine-tuned encoder-decoder architecture that integrates LLM-assisted compression with structured QA guidance. On standard benchmarks, it achieves a ROUGE-L score of 0.4437—surpassing the state-of-the-art REFACTOR by +2.17%—marking the first approach to jointly achieve high accuracy and strong scalability for long-table settings. This work establishes a novel paradigm for controllable, precise summarization under lengthy tabular conditions.

Technology Category

• Natural Language Processing: Summarization
• Machine Learning: Large Multimodal Models (LMMs)
• Data Mining & Knowledge Management: Intelligent Query Processing

Application Category

• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Query-focused tabular summarization is an emerging task in table-to-text generation that synthesizes a summary response from tabular data based on user queries. Traditional transformer-based approaches face challenges due to token limitations and the complexity of reasoning over large tables. To address these challenges, we introduce DETQUS (Decomposition-Enhanced Transformers for QUery-focused Summarization), a system designed to improve summarization accuracy by leveraging tabular decomposition alongside a fine-tuned encoder-decoder model. DETQUS employs a large language model to selectively reduce table size, retaining only query-relevant columns while preserving essential information. This strategy enables more efficient processing of large tables and enhances summary quality. Our approach, equipped with table-based QA model Omnitab, achieves a ROUGE-L score of 0.4437, outperforming the previous state-of-the-art REFACTOR model (ROUGE-L: 0.422). These results highlight DETQUS as a scalable and effective solution for query-focused tabular summarization, offering a structured alternative to more complex architectures.

Problem

• Improves query-focused tabular summarization accuracy
• Addresses token limitations and table complexity challenges
• Enhances summary quality via selective table reduction

Innovation

• Decomposition-Enhanced Transformers for efficient summarization
• Selective table reduction using large language models
• Integration with Omnitab for improved ROUGE-L scores