DataFactory: Collaborative Multi-Agent Framework for Advanced Table Question Answering
arXiv:2603.09152v1 Announce Type: new Abstract: 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
arXiv:2603.09152v1 Announce Type: new Abstract: 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.
Executive Summary
This article introduces DataFactory, a multi-agent framework designed to address the limitations of existing large language model (LLM) approaches in Table Question Answering (TableQA). The framework employs specialized team coordination and automated knowledge transformation to enable systematic decomposition of complex queries into structured and relational reasoning tasks. Results across three datasets demonstrate significant improvements in accuracy, with gains of up to 23.9% over baselines. 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.
Key Points
- ▸ DataFactory is a multi-agent framework designed to address limitations of existing LLM approaches in TableQA
- ▸ The framework employs specialized team coordination and automated knowledge transformation
- ▸ Results demonstrate significant improvements in accuracy, with gains of up to 23.9% over baselines
Merits
Addressing Critical Limitations
DataFactory effectively addresses limitations of existing LLM approaches, including context length constraints, hallucination issues, and single-agent architectures
Improved Accuracy
The framework achieves significant improvements in accuracy, with gains of up to 23.9% over baselines
Scalability and Flexibility
DataFactory offers design guidelines for multi-agent collaboration and a practical platform for enterprise data analysis
Demerits
Complexity
The framework's multi-agent architecture and automated knowledge transformation may introduce complexity and require significant computational resources
Limited Evaluation
The article's evaluation is limited to three datasets, and it is unclear whether the results generalize to other domains or datasets
Expert Commentary
The introduction of DataFactory is a significant advancement in the field of Table Question Answering. The framework's ability to address critical limitations of existing LLM approaches, such as context length constraints and hallucination issues, is a major step forward. However, the complexity of the framework and the limited evaluation may limit its adoption and generalizability. Nevertheless, the framework's potential to improve accuracy and scalability makes it an exciting development in the field. As the field of TableQA continues to evolve, DataFactory is likely to play a significant role in shaping the future of data analysis and question answering.
Recommendations
- ✓ Further evaluation of the framework across multiple datasets and domains is necessary to establish its generalizability and robustness
- ✓ The development of more efficient and scalable algorithms for automated knowledge transformation is crucial to improve the framework's performance
