TableRAG: A Retrieval Augmented Generation Framework for Heterogeneous Document Reasoning
By: Xiaohan Yu, Pu Jian, Chong Chen
Potential Business Impact:
Helps computers answer questions from mixed text and tables.
Retrieval-Augmented Generation (RAG) has demonstrated considerable effectiveness in open-domain question answering. However, when applied to heterogeneous documents, comprising both textual and tabular components, existing RAG approaches exhibit critical limitations. The prevailing practice of flattening tables and chunking strategies disrupts the intrinsic tabular structure, leads to information loss, and undermines the reasoning capabilities of LLMs in multi-hop, global queries. To address these challenges, we propose TableRAG, an hybrid framework that unifies textual understanding and complex manipulations over tabular data. TableRAG iteratively operates in four steps: context-sensitive query decomposition, text retrieval, SQL programming and execution, and compositional intermediate answer generation. We also develop HeteQA, a novel benchmark designed to evaluate the multi-hop heterogeneous reasoning capabilities. Experimental results demonstrate that TableRAG consistently outperforms existing baselines on both public datasets and our HeteQA, establishing a new state-of-the-art for heterogeneous document question answering. We release TableRAG at https://github.com/yxh-y/TableRAG/tree/main.
Similar Papers
Mixture-of-RAG: Integrating Text and Tables with Large Language Models
Information Retrieval
Helps computers understand text and tables together.
GTR: Graph-Table-RAG for Cross-Table Question Answering
Computation and Language
Helps computers answer questions from many tables.
SRAG: Structured Retrieval-Augmented Generation for Multi-Entity Question Answering over Wikipedia Graph
Computation and Language
Helps computers answer questions from many papers.