CFMS: A Coarse-to-Fine Multimodal Synthesis Framework for Enhanced Tabular Reasoning

TL;DR

CFMS introduces a hierarchical two-stage framework that combines multimodal perception and symbolic reasoning to improve tabular data understanding for question answering and fact verification.

Contribution

It proposes a novel coarse-to-fine paradigm that decouples visual perception from symbolic reasoning, enhancing robustness and efficiency in tabular reasoning tasks.

Findings

CFMS achieves competitive accuracy on WikiTQ and TabFact benchmarks.

The framework is robust with large tables and smaller models.

Extensive experiments validate its effectiveness and generalizability.

Abstract

Reasoning over tabular data is a crucial capability for tasks like question answering and fact verification, as it requires models to comprehend both free-form questions and semi-structured tables. However, while methods like Chain-of-Thought (CoT) introduce reasoning chains, purely symbolic methodes are inherently limited by their blindness to holistic visual patterns. To address this, we propose the Coarse-to-Fine Multimodal Synthesis framework (CFMS), a novel two-stage paradigm that hierarchically decouples high-level visual perception from granular symbolic reasoning. In the Coarse Stage, CFMS leverages the Multimodal Large Language Models (MLLMs) to perform a one-time synthesis of a multi-perspective knowledge tuple. This tuple subsequently serves as a dynamic reasoning map to guide the fine stage, where a symbolic engine executes a targeted and efficient sequence of iterative operations over the table. Extensive experiments on the WikiTQ and TabFact benchmarks demonstrate that CFMS achieves competitive accuracy. The framework exhibits particular robustness when handling large tables and when instantiated with smaller backbone models, validating its effectiveness and generalizability.