Theorem-Validated Reverse Chain-of-Thought Problem Generation for Geometric Reasoning

TL;DR

This paper introduces a two-stage theorem-validated reverse chain-of-thought framework to generate diverse, precise geometric reasoning data, significantly improving model understanding and accuracy in geometric reasoning tasks.

Contribution

The proposed TR-CoT framework uniquely synthesizes theorem-grounded diagrams and refines question-answer pairs through reverse reasoning, enhancing geometric reasoning data quality.

Findings

Increases logical consistency by 24.5%.

Surpasses baselines in MathVista and GeoQA by 10.1% and 4.7%.

Outperforms advanced models like GPT-4o.

Abstract

Large Multimodal Models (LMMs) face limitations in geometric reasoning due to insufficient Chain of Thought (CoT) image-text training data. While existing approaches leverage template-based or LLM-assisted methods for geometric CoT data creation, they often face challenges in achieving both diversity and precision. To bridge this gap, we introduce a two-stage Theorem-Validated Reverse Chain-of-Thought Reasoning Synthesis (TR-CoT) framework. The first stage, TR-Engine, synthesizes theorem-grounded geometric diagrams with structured descriptions and properties. The second stage, TR-Reasoner, employs reverse reasoning to iteratively refine question-answer pairs by cross-validating geometric properties and description fragments. Our approach expands theorem-type coverage, corrects long-standing misunderstandings, and enhances geometric reasoning. Fine-grained CoT improves theorem understanding and increases logical consistency by 24.5%. Our best models surpass the baselines in MathVista and GeoQA by 10.1% and 4.7%, outperforming advanced closed-source models like GPT-4o.