ST-Raptor: An Agentic System for Semi-Structured Table QA

2026-02-03

AI Summary

This work addresses the challenges of semi-structured table question answering, which include precise extraction of cell content and positional information, recovery of implicit logical structures, and semantic relationships. Conventional approaches often suffer from information loss or inaccurate answers due to format conversion during preprocessing. To overcome these limitations, the paper proposes a novel system architecture that integrates a visual editing interface, tree-based hierarchical modeling, and multi-agent collaboration, all orchestrated by large language model–driven agents to enable end-to-end query parsing and interactive question answering. By circumventing traditional preprocessing steps that degrade information fidelity, the proposed method significantly enhances both the accuracy of complex table understanding and the user interaction experience, achieving state-of-the-art performance in answer accuracy and system usability across multiple benchmark and real-world datasets.

Technology Category

• Natural Language Processing: Question Answering
• Data Mining & Knowledge Management: Intelligent Query Processing
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Web question answering
• Web Mining and Content Analysis: Community question answering

Abstract

Semi-structured table question answering (QA) is a challenging task that requires (1) precise extraction of cell contents and positions and (2) accurate recovery of key implicit logical structures, hierarchical relationships, and semantic associations encoded in table layouts. In practice, such tables are often interpreted manually by human experts, which is labor-intensive and time-consuming. However, automating this process remains difficult. Existing Text-to-SQL methods typically require converting semi-structured tables into structured formats, inevitably leading to information loss, while approaches like Text-to-Code and multimodal LLM-based QA struggle with complex layouts and often yield inaccurate answers. To address these limitations, we present ST-Raptor, an agentic system for semi-structured table QA. ST-Raptor offers an interactive analysis environment that combines visual editing, tree-based structural modeling, and agent-driven query resolution to support accurate and user-friendly table understanding. Experimental results on both benchmark and real-world datasets demonstrate that ST-Raptor outperforms existing methods in both accuracy and usability. The code is available at https://github.com/weAIDB/ST-Raptor, and a demonstration video is available at https://youtu.be/9GDR-94Cau4.

Problem

• semi-structured table
• question answering
• logical structure recovery
• layout understanding
• information extraction

Innovation

• agentic system
• semi-structured table QA
• tree-based structural modeling
• visual editing
• implicit logical recovery