Reproduce “Generating Analytic Specifications for Data Visualization from Natural Language Queries using Large Language Models” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Subham Sah, Rishab Mitra, Arpit Narechania, Alex Endert, John Stasko, Wenwen Dou

NL4DV-LLM uses GPT-4 to turn natural-language data queries into structured interpretations and Vega-Lite visualizations. Its output identifies matched attributes, inferred analytic tasks, and chart recommendations, and can map attributes to query phrases. In a manual evaluation of 740 queries, it was judged accurate on 87.02%, versus 64.05% for NL4DV, across three dataset domains.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.