用多智能体协作解决表格问答中的幻觉和复杂推理难题
DataFactory: Collaborative Multi-Agent Framework for Advanced Table Question Answering
- 设计多智能体框架，分工处理数据、知识图谱与推理协调
- 在三个数据集上准确率提升超20%，显著优于单智能体方法
- 适合企业级数据查询与需要高可靠性的复杂分析场景
表格问答（TableQA）支持自然语言与结构化表格数据的交互。现有大模型方法存在三大瓶颈：上下文长度限制导致数据处理能力不足、幻觉问题影响答案可靠性、单智能体架构难以应对涉及语义关系与多跳逻辑的复杂推理。本文提出DataFactory，一个基于多智能体协作的框架，通过专业化团队协同与自动知识转化机制克服上述问题。框架包含采用ReAct范式的数据领导者，以及专责数据库与知识图谱的团队，实现复杂查询向结构化与关联推理任务的系统性分解。我们形式化定义了数据到知识图谱的映射函数 T:D × S × R → G，并引入基于自然语言的协商机制，支持灵活跨智能体讨论与自适应规划，提升协调鲁棒性。同时应用上下文工程策略，融合历史模式与领域知识，降低幻觉并提高查询准确性。在TabFact、WikiTableQuestions和FeTaQA三个数据集上，使用五家厂商的八种LLM进行测试，结果表明本方法持续提升性能：相比基线，准确率分别提升20.2%（TabFact）和23.9%（WikiTQ），效应量均大于1。团队协作效果优于单团队变体（TabFact+5.5%，WikiTQ+14.4%，FeTaQA ROUGE-2+17.1%）。该框架为多智能体协作提供设计指南，并构建了面向企业数据分析的实用平台，整合结构化查询与图式知识表示。
原文摘要 · Abstract (English)
Table Question Answering (TableQA) enables natural language interaction with structured tabular data. However, existing large language model (LLM) approaches face critical limitations: context length constraints that restrict data handling capabilities, hallucination issues that compromise answer reliability, and single-agent architectures that struggle with complex reasoning scenarios involving semantic relationships and multi-hop logic. This paper introduces DataFactory, a multi-agent framework that addresses these limitations through specialized team coordination and automated knowledge transformation. The framework comprises a Data Leader employing the ReAct paradigm for reasoning orchestration, together with dedicated Database and Knowledge Graph teams, enabling the systematic decomposition of complex queries into structured and relational reasoning tasks. We formalize automated data-to-knowledge graph transformation via the mapping function T:D x S x R -> G, and implement natural language-based consultation that - unlike fixed workflow multi-agent systems - enables flexible inter-agent deliberation and adaptive planning to improve coordination robustness. We also apply context engineering strategies that integrate historical patterns and domain knowledge to reduce hallucinations and improve query accuracy. Across TabFact, WikiTableQuestions, and FeTaQA, using eight LLMs from five providers, results show consistent gains. Our approach improves accuracy by 20.2% (TabFact) and 23.9% (WikiTQ) over baselines, with significant effects (Cohen's d > 1). Team coordination also outperforms single-team variants (+5.5% TabFact, +14.4% WikiTQ, +17.1% FeTaQA ROUGE-2). The framework offers design guidelines for multi-agent collaboration and a practical platform for enterprise data analysis through integrated structured querying and graph-based knowledge representation.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。