Reproduce “EmphasisChecker: A Tool for Guiding Chart and Caption Emphasis” 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
Dae Hyun Kim, Seulgi Choi, Juho Kim, Vidya Setlur, Maneesh Agrawala
EMPHASISCHECKER, an interactive authoring tool, helps authors align visual emphasis in time-series line charts with the features highlighted in their captions. This tool enhances the effectiveness of data communication by identifying mismatches and factual errors, with its prominent feature detector outperforming existing methods and users finding its guidance useful for improving message clarity.
Key insights
- The e-persistence technique effectively identifies visually prominent features in time-series line charts, achieving performance comparable to human perception and outperforming prior methods.
- A multi-step NLP pipeline can successfully extract complex temporal and descriptive data references from natural language captions and accurately map them to specific visual chart features.
- Interactive, "checker-style" guidance for visual-textual alignment is highly valued by authors and improves communication effectiveness without compromising author control.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.