FT-RAG: A Fine-grained Retrieval-Augmented Generation Framework for Complex Table Reasoning

TL;DR

FT-RAG introduces a fine-grained, graph-based retrieval-augmented generation framework that significantly improves complex table reasoning and factual grounding in LLMs.

Contribution

The paper proposes FT-RAG, a novel fine-grained retrieval framework with a new benchmark, achieving state-of-the-art results in complex multi-table reasoning tasks.

Findings

FT-RAG achieves 23.5 ext% and 59.2 ext% improvements in table and cell hit rates.

Generation accuracy recall increases by 62.2 ext%.

FT-RAG outperforms existing baselines across all evaluated metrics.

Abstract

Retrieval-Augmented Generation (RAG) enhances Large Language Models (LLMs) by grounding responses in external knowledge during inference. However, conventiona RAG systems under-perform on structured tabular data, largely due to coarse retrieval granularity and insufficient table semantic comprehension. To address these limitations, we introduce FT-RAG, a fine-grained framework that employs knowledge association by decomposing tables into entry-level semantic units to construct a structured graph. FT-RAG employs a structural neighbor expansion mechanism to find semantically connected entities during graph retrieval, followed by multi-modal fusion to consolidate the context of table retrieval results. Further, to address the scarcity of specialized datasets in this domain, we introduce Multi-Table-RAG-Lib, a benchmark comprising 9870 QA pairs with high complexity and difficulty, curated to demand multi-table integration and text-table information fusion for reasoning. FT-RAG surpasses top-performing baselines across all metrics, achieving a 23.5\% and 59.2\% improvement in table-level and cell-level Hit Rates, respectively. Generation performance also sees a remarkable 62.2\% increase in exact value accuracy recall. These metrics verify the framework's effectiveness in factual grounding across both pure tabular and heterogeneous table-text contexts. Therefore, our method establishes a new state-of-the-art performance for complex reasoning over mixed-modality documents.