Papers
arxiv:2406.17294

Math-LLaVA: Bootstrapping Mathematical Reasoning for Multimodal Large Language Models

Published on Jun 25
· Submitted by Zhiqiang007 on Jun 27
Authors:
,
,
,
,
,

Abstract

Large language models (LLMs) have demonstrated impressive reasoning capabilities, particularly in textual mathematical problem-solving. However, existing open-source image instruction fine-tuning datasets, containing limited question-answer pairs per image, do not fully exploit visual information to enhance the multimodal mathematical reasoning capabilities of Multimodal LLMs (MLLMs). To bridge this gap, we address the lack of high-quality, diverse multimodal mathematical datasets by collecting 40K high-quality images with question-answer pairs from 24 existing datasets and synthesizing 320K new pairs, creating the MathV360K dataset, which enhances both the breadth and depth of multimodal mathematical questions. We introduce Math-LLaVA, a LLaVA-1.5-based model fine-tuned with MathV360K. This novel approach significantly improves the multimodal mathematical reasoning capabilities of LLaVA-1.5, achieving a 19-point increase and comparable performance to GPT-4V on MathVista's minitest split. Furthermore, Math-LLaVA demonstrates enhanced generalizability, showing substantial improvements on the MMMU benchmark. Our research highlights the importance of dataset diversity and synthesis in advancing MLLMs' mathematical reasoning abilities. The code and data are available at: https://github.com/HZQ950419/Math-LLaVA.

Community

Paper author Paper submitter
edited Jun 27

We introduce Math-LLaVA, a LLaVA-1.5-based model fine-tuned with MathV360K. This novel approach significantly improves the multimodal mathematical reasoning capabilities of LLaVA-1.5, achieving a 19-point increase and comparable performance to GPT-4V on MathVista's minitest split. And Math-LLaVA achieves 15.69% accuracy on MathVision benchmark, outperforming Qwen-VL-Max (15.59%).

GitHub: https://github.com/HZQ950419/Math-LLaVA
Model: https://huggingface.co/Zhiqiang007/Math-LLaVA
Data: https://huggingface.co/Zhiqiang007/MathV360K

Sign up or log in to comment

Models citing this paper 1

Datasets citing this paper 1

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2406.17294 in a Space README.md to link it from this page.

Collections including this paper 6