TART: An Open-Source Tool-Augmented Framework for Explainable Table-based Reasoning

Current Large Language Models (LLMs) exhibit limited ability to understand table structures and to apply precise numerical reasoning, which is crucial for tasks such as table question answering and table-based fact verification. To address these challenges, we introduce our Tool-Augmented Reasoning framework for Tables (TART), which integrates LLMs with specialized tools. TART contains three key components: a table formatter to ensure accurate data representation, a tool maker to develop specific computational tools, and an explanation generator to maintain explainability. We also present the TOOLTAB dataset, a new benchmark designed specifically for training LLMs in table–tool integration. Our experiments indicate that TART achieves substantial improvements over existing methods (e.g., Chain-of-Thought) by improving both the precision of data processing and the clarity of the reasoning process. Notably, TART paired with CodeLlama achieves 90.0% of the accuracy of the closed-sourced LLM GPT-3.5-turbo, highlighting its robustness in diverse real-world scenarios. Both code and data are openly available at https://github.com/XinyuanLu00/TART.

Citation

X. Lu, L. Pan, Y. Ma, P. Nakov, M.-Y. Kan, "TART: An Open-Source Tool-Augmented Framework for Explainable Table-based Reasoning," 2025, pp. 4323-4339.

Source

Proceedings of the Conference Findings, NAACL 2025

Conference

2025 Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics

Source

2025 Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics

Publisher

Association for Computational Linguistics (ACL)

DOI

10.18653/v1/2025.findings-naacl.244

Additional links

https://aclanthology.org/2025.findings-naacl.244/