Understanding Graphical Perception in Data Visualization through Zero-shot Prompting of Vision-Language Models
This work addresses the problem of understanding VLM alignment with human chart comprehension for researchers and practitioners in data visualization, enabling broader applications like designing and evaluating visualizations, though it is incremental in establishing foundational insights.
The paper evaluated zero-shot prompting of Vision-Language Models (VLMs) on graphical perception tasks with known human performance, finding that VLMs perform similarly to humans in specific task and style combinations, indicating potential for modeling human performance.
Vision Language Models (VLMs) have been successful at many chart comprehension tasks that require attending to both the images of charts and their accompanying textual descriptions. However, it is not well established how VLM performance profiles map to human-like behaviors. If VLMs can be shown to have human-like chart comprehension abilities, they can then be applied to a broader range of tasks, such as designing and evaluating visualizations for human readers. This paper lays the foundations for such applications by evaluating the accuracy of zero-shot prompting of VLMs on graphical perception tasks with established human performance profiles. Our findings reveal that VLMs perform similarly to humans under specific task and style combinations, suggesting that they have the potential to be used for modeling human performance. Additionally, variations to the input stimuli show that VLM accuracy is sensitive to stylistic changes such as fill color and chart contiguity, even when the underlying data and data mappings are the same.