ChartX & ChartVLM: A Versatile Benchmark and Foundation Model for Complicated Chart Reasoning

TL;DR

This paper introduces ChartX, a comprehensive benchmark for multi-modal models in chart reasoning, and ChartVLM, a new model that outperforms existing models in interpreting and reasoning with complex visual charts.

Contribution

The paper presents a new multi-modal evaluation set for charts and a novel model, ChartVLM, that excels in chart reasoning tasks, surpassing existing models and approaching GPT-4V performance.

Findings

ChartVLM outperforms other models on ChartX benchmark.

ChartX covers 18 chart types and 7 tasks, providing a comprehensive evaluation.

ChartVLM achieves results comparable to GPT-4V.

Abstract

Recently, many versatile Multi-modal Large Language Models (MLLMs) have emerged continuously. However, their capacity to query information depicted in visual charts and engage in reasoning based on the queried contents remains under-explored. In this paper, to comprehensively and rigorously benchmark the ability of the off-the-shelf MLLMs in the chart domain, we construct ChartX, a multi-modal evaluation set covering 18 chart types, 7 chart tasks, 22 disciplinary topics, and high-quality chart data. Besides, we develop ChartVLM to offer a new perspective on handling multi-modal tasks that strongly depend on interpretable patterns, such as reasoning tasks in the field of charts or geometric images. We evaluate the chart-related ability of mainstream MLLMs and our ChartVLM on the proposed ChartX evaluation set. Extensive experiments demonstrate that ChartVLM surpasses both versatile and chart-related large models, achieving results comparable to GPT-4V. We believe that our study can pave the way for further exploration in creating a more comprehensive chart evaluation set and developing more interpretable multi-modal models. Both ChartX and ChartVLM are available at: https://github.com/Alpha-Innovator/ChartVLM