ChartSumm: A Comprehensive Benchmark for Automatic Chart Summarization of Long and Short Summaries
This provides a comprehensive benchmark for researchers working on chart summarization, though it is incremental as it focuses on dataset creation.
The authors introduced ChartSumm, a large-scale benchmark dataset of 84,363 charts with metadata and descriptions for automatic chart summarization, and found that baseline models generate fluent summaries but often hallucinate, miss key data, or incorrectly explain trends.
Automatic chart to text summarization is an effective tool for the visually impaired people along with providing precise insights of tabular data in natural language to the user. A large and well-structured dataset is always a key part for data driven models. In this paper, we propose ChartSumm: a large-scale benchmark dataset consisting of a total of 84,363 charts along with their metadata and descriptions covering a wide range of topics and chart types to generate short and long summaries. Extensive experiments with strong baseline models show that even though these models generate fluent and informative summaries by achieving decent scores in various automatic evaluation metrics, they often face issues like suffering from hallucination, missing out important data points, in addition to incorrect explanation of complex trends in the charts. We also investigated the potential of expanding ChartSumm to other languages using automated translation tools. These make our dataset a challenging benchmark for future research.