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

2026-04-13

AI Summary

This work addresses the limited ability of existing symbolic approaches to effectively capture the holistic visual patterns of tables. To overcome this challenge, the authors propose a coarse-to-fine two-stage reasoning framework: first, leveraging multimodal large language models (MLLMs) to generate knowledge tuples from multiple visual perspectives, thereby enabling coarse-grained table understanding; then guiding a symbolic reasoning engine to perform fine-grained, iterative table operations. The approach innovatively decouples visual perception from symbolic reasoning in a hierarchical manner and constructs a dynamic reasoning map, significantly enhancing adaptability to both large-scale tables and smaller backbone models. Competitive accuracy is achieved on the WikiTQ and TabFact benchmarks, with particularly robust performance on complex and large-scale tabular data.

Technology Category

• Computer Vision: Visual Reasoning & Symbolic Representations
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning
• Cognitive Modeling & Cognitive Systems: Symbolic Representations

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

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.

Problem

• tabular reasoning
• multimodal synthesis
• visual patterns
• symbolic reasoning
• question answering

Innovation

• Coarse-to-Fine
• Multimodal Synthesis
• Tabular Reasoning
• Multimodal Large Language Models
• Symbolic Reasoning