VisualSphinx: Large-Scale Synthetic Vision Logic Puzzles for RL

TL;DR

VisualSphinx introduces a large-scale synthetic dataset for visual logical reasoning, significantly improving the reasoning capabilities of vision language models across multiple domains.

Contribution

It presents a novel rule-to-image synthesis pipeline and a large-scale dataset to enhance multimodal reasoning in vision language models.

Findings

VLM trained on VisualSphinx shows improved logical reasoning performance.

Enhanced reasoning extends to algebraic, arithmetic, and geometry tasks.

Dataset improves logical coherence and readability in model training.

Abstract

Vision language models (VLMs) are expected to perform effective multimodal reasoning and make logically coherent decisions, which is critical to tasks such as diagram understanding and spatial problem solving. However, current VLM reasoning lacks large-scale and well-structured training datasets. To bridge this gap, we propose VisualSphinx, a first-of-its-kind large-scale synthetic visual logical reasoning training data. To tackle the challenge of image synthesis with grounding answers, we propose a rule-to-image synthesis pipeline, which extracts and expands puzzle rules from seed questions and generates the code of grounding synthesis image synthesis for puzzle sample assembly. Experiments demonstrate that VLM trained using GRPO on VisualSphinx benefit from logical coherence and readability of our dataset and exhibit improved performance on logical reasoning tasks. The enhanced reasoning capabilities developed from VisualSphinx also benefit other reasoning tasks such as algebraic reasoning, arithmetic reasoning and geometry reasoning.