Beyond Natural Language Plans: Structure-Aware Planning for Query-Focused Table Summarization

2025-07-30

AI Summary

Query-focused table summarization relies on natural language (NL) plans, but their inherent ambiguity and lack of structure severely limit SQL generation accuracy and scalability to multi-table scenarios. To address this, we propose SPaGe—a novel structured reasoning paradigm comprising three stages: *structured planning*, *graph-driven execution*, and *summary generation*. Our key contributions are: (1) TaSoF, a multi-agent-inspired structured planner that explicitly models cross-table dependencies and encodes NL plans as executable directed cyclic graphs amenable to parallel execution; and (2) a formal graph-based execution mechanism that robustly maps structured plans to executable SQL. Evaluated on three public benchmarks, SPaGe achieves state-of-the-art performance in both single- and multi-table settings, while significantly improving interpretability, robustness, and scalability of query-to-SQL reasoning.

Technology Category

• Multiagent Systems: Multiagent Planning
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Query-focused table summarization requires complex reasoning, often approached through step-by-step natural language (NL) plans. However, NL plans are inherently ambiguous and lack structure, limiting their conversion into executable programs like SQL and hindering scalability, especially for multi-table tasks. To address this, we propose a paradigm shift to structured representations. We introduce a new structured plan, TaSoF, inspired by formalism in traditional multi-agent systems, and a framework, SPaGe, that formalizes the reasoning process in three phases: 1) Structured Planning to generate TaSoF from a query, 2) Graph-based Execution to convert plan steps into SQL and model dependencies via a directed cyclic graph for parallel execution, and 3) Summary Generation to produce query-focused summaries. Our method explicitly captures complex dependencies and improves reliability. Experiments on three public benchmarks show that SPaGe consistently outperforms prior models in both single- and multi-table settings, demonstrating the advantages of structured representations for robust and scalable summarization.

Problem

• Address ambiguity in natural language plans for table summarization
• Improve scalability of multi-table query-focused summarization tasks
• Enhance reliability by capturing complex dependencies in structured plans

Innovation

• Structured plan TaSoF replaces ambiguous NL plans
• Graph-based execution enables parallel SQL conversion
• Three-phase framework SPaGe enhances reliability and scalability