From the Least to the Most: Building a Plug-and-Play Visual Reasoner via Data Synthesis

TL;DR

This paper introduces a novel data synthesis method for multi-step visual reasoning, enabling the enhancement of vision-language models through a plug-and-play visual reasoner trained on automatically generated reasoning data.

Contribution

It proposes a least-to-most visual reasoning paradigm and a cost-effective data synthesis approach to generate multi-step reasoning examples for training visual reasoners.

Findings

Significant improvement in VQA benchmarks across multiple models.

Reproducible and cost-efficient data synthesis process.

Enhanced reasoning abilities in existing vision-language models.

Abstract

We explore multi-step reasoning in vision-language models (VLMs). The problem is challenging, as reasoning data consisting of multiple steps of visual and language processing are barely available. To overcome the challenge, we first introduce a least-to-most visual reasoning paradigm, which interleaves steps of decomposing a question into sub-questions and invoking external tools for resolving sub-questions. Based on the paradigm, we further propose a novel data synthesis approach that can automatically create questions and multi-step reasoning paths for an image in a bottom-up manner. Our approach divides the complex synthesis task into a few simple sub-tasks, and (almost entirely) relies on open-sourced models to accomplish the sub-tasks. Therefore, the entire synthesis process is reproducible and cost-efficient, and the synthesized data is quality guaranteed. With the approach, we construct $50$k visual reasoning examples. Then, we develop a visual reasoner through supervised fine-tuning, which is capable of generally enhancing the reasoning abilities of a wide range of existing VLMs in a plug-and-play fashion. Extensive experiments indicate that the visual reasoner can consistently and significantly improve four VLMs on four VQA benchmarks. Our code and dataset are available at https://github.com/steven-ccq/VisualReasoner.