ChartThinker: A Contextual Chain-of-Thought Approach to Optimized Chart Summarization

TL;DR

ChartThinker introduces a novel chain-of-thought approach for chart summarization, leveraging a large dataset and context retrieval to enhance logical coherence and accuracy in data visualization summaries.

Contribution

The paper presents a new dataset and a chain-of-thought based method, significantly improving chart summarization performance over existing models.

Findings

Outperforms 8 state-of-the-art models across 7 metrics

Large-scale dataset improves visual-language matching

Chain-of-thought approach enhances reasoning in summaries

Abstract

Data visualization serves as a critical means for presenting data and mining its valuable insights. The task of chart summarization, through natural language processing techniques, facilitates in-depth data analysis of charts. However, there still are notable deficiencies in terms of visual-language matching and reasoning ability for existing approaches. To address these limitations, this study constructs a large-scale dataset of comprehensive chart-caption pairs and fine-tuning instructions on each chart. Thanks to the broad coverage of various topics and visual styles within this dataset, better matching degree can be achieved from the view of training data. Moreover, we propose an innovative chart summarization method, ChartThinker, which synthesizes deep analysis based on chains of thought and strategies of context retrieval, aiming to improve the logical coherence and accuracy of the generated summaries. Built upon the curated datasets, our trained model consistently exhibits superior performance in chart summarization tasks, surpassing 8 state-of-the-art models over 7 evaluation metrics. Our dataset and codes are publicly accessible.