TART: An Open-Source Tool-Augmented Framework for Explainable Table-based Reasoning
This addresses table-based reasoning challenges for AI applications like question answering and fact verification, representing an incremental improvement with a novel method for a known bottleneck.
The paper tackles the problem of limited table structure understanding and numerical reasoning in Large Language Models (LLMs) by introducing TART, a tool-augmented framework, which achieves 90.0% accuracy compared to GPT-3.5-turbo on table-based tasks.
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 (TQA) and table-based fact verification (TFV). 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. All the code and data are available at https://github.com/XinyuanLu00/TART.