VisEval: A Benchmark for Data Visualization in the Era of Large Language Models
This work addresses a gap in benchmarking for data visualization with LLMs, which is incremental as it provides a tool for evaluation rather than a new method.
The authors tackled the lack of a comprehensive benchmark for evaluating large language models (LLMs) in generating visualizations from natural language by proposing VisEval, a new NL2VIS benchmark that includes a dataset of 2,524 queries across 146 databases and an automated evaluation methodology, revealing prevalent challenges in LLM capabilities.
Translating natural language to visualization (NL2VIS) has shown great promise for visual data analysis, but it remains a challenging task that requires multiple low-level implementations, such as natural language processing and visualization design. Recent advancements in pre-trained large language models (LLMs) are opening new avenues for generating visualizations from natural language. However, the lack of a comprehensive and reliable benchmark hinders our understanding of LLMs' capabilities in visualization generation. In this paper, we address this gap by proposing a new NL2VIS benchmark called VisEval. Firstly, we introduce a high-quality and large-scale dataset. This dataset includes 2,524 representative queries covering 146 databases, paired with accurately labeled ground truths. Secondly, we advocate for a comprehensive automated evaluation methodology covering multiple dimensions, including validity, legality, and readability. By systematically scanning for potential issues with a number of heterogeneous checkers, VisEval provides reliable and trustworthy evaluation outcomes. We run VisEval on a series of state-of-the-art LLMs. Our evaluation reveals prevalent challenges and delivers essential insights for future advancements.