What problem does it solve?
Manually creating publication-quality figures and tables for academic papers is time-consuming, requires specialized plotting expertise, and often produces inconsistent formatting that fails to meet top-tier conference submission standards.
Core Features & Use Cases
- Automated data visualization: Generates line plots, bar charts, scatter plots, heatmaps, box/violin plots, and multi-panel subfigures directly from experiment JSON or CSV data.
- LaTeX table generation: Creates print-ready comparison tables for method benchmarks and ablation studies, formatted for direct inclusion in LaTeX manuscripts.
- Consistent publication styling: Enforces vector PDF output, conference-matched serif fonts, colorblind-safe color palettes, and proper sizing for single-column or full-width placement.
- Use case: For a MICCAI 2025 vertebrae segmentation paper, use this skill to automatically generate training loss curves, state-of-the-art comparison bar charts, and ablation study tables from experiment result files, all formatted to meet conference requirements.
Quick Start
Use the paper-figure skill to generate all publication-ready figures and LaTeX include snippets from the experiment data in your project's figures/ directory and the figure plan in PAPER_PLAN.md.