table-generation
lingzhi227/agent-research-skills
Generate publication-quality LaTeX tables from experimental results with booktabs styling and bold best results.
What is table-generation?
Converts JSON/CSV experimental data into LaTeX tables formatted for academic papers. Supports comparison, ablation, and descriptive table types with automatic best-result highlighting, multi-row layouts, and proper captions. Use when preparing result tables for research papers or technical documents.
- Convert JSON/CSV data to booktabs-styled LaTeX tables
- Automatically bold best results in each column with optional second-best underlining
- Generate comparison tables (methods × metrics), ablation tables (variants × metrics), and descriptive tables
- Support multi-row layouts with \multirow for method categories and component grouping
- Include mean ± std formatting and p-value significance stars
- Add captions, labels, and table notes via threeparttable environment
How to install table-generation
npx skills add https://github.com/lingzhi227/agent-research-skills --skill table-generation- Python with pandas or JSON/CSV parsing capability
- LaTeX packages: booktabs, multirow, multicol, threeparttable
- Experimental results in JSON or CSV format
How to use table-generation
- 1.Prepare your experimental data in JSON or CSV format with methods/variants as rows and metrics as columns
- 2.Run the table-generation script with --input pointing to your data file, --type specifying table kind (comparison/ablation/descriptive/custom), and --caption with your table title
- 3.Specify --bold-best max or min depending on metric direction, and optionally add --underline-second for second-best highlighting
- 4.Copy the generated LaTeX code into your document preamble (ensure required packages are included) and paste the table environment into your results section
- 5.Verify all numbers match your source data and adjust column alignment or notes as needed
Use cases
- Create main results comparison table showing baseline and proposed method performance across datasets
- Generate ablation study table documenting impact of removing each model component
- Produce dataset statistics or hyperparameter summary table for appendix
- Format multi-dataset evaluation table with methods as rows and dataset×metric combinations as columns
- Build custom table layouts with user-specified structure and content
- Researchers writing machine learning or NLP papers
- Scientists preparing experimental results for publication
- Graduate students creating thesis tables and figures
- Anyone generating LaTeX tables from structured experimental data
table-generation FAQ
JSON files and CSV files. Data should have methods/variants as rows and metrics/datasets as columns. The script parses both formats automatically.
Yes, use the --underline-second flag to underline second-best results in each column. Best results are always bolded by default.
Use the --significance flag to add p-value stars (*, **, ***) to results when p-values are included in your data.
Your document preamble must include: \usepackage{booktabs}, \usepackage{multirow}, \usepackage{multicol}, and \usepackage{threeparttable}.
Yes, the script generates table* environments for wide tables. Use this for multi-dataset layouts that span both columns.
Full instructions (SKILL.md)
Source of truth, from lingzhi227/agent-research-skills.
name: table-generation description: Generate publication-quality LaTeX tables from experimental results. Convert JSON/CSV data to booktabs-styled tables with bold best results, multi-row layouts, and proper captions. Use when creating result tables, comparison tables, or ablation tables for papers. argument-hint: [data-source]
Table Generation
Convert experimental results into publication-ready LaTeX tables.
Input
$0— Table type:comparison,ablation,descriptive,custom$1— Data source: JSON file, CSV file, or inline data
Scripts
Generate LaTeX table from JSON/CSV
python ~/.claude/skills/table-generation/scripts/results_to_table.py \
--input results.json --type comparison \
--bold-best max --caption "Performance comparison" \
--label tab:main_results
Supports: comparison, ablation, descriptive, multi-dataset table types.
Additional flags: --type multi-dataset for methods x datasets x metrics layout, --significance for p-value stars, --underline-second for second-best results.
References
- LaTeX table templates and examples:
~/.claude/skills/table-generation/references/table-templates.md
Table Types
comparison — Main results table
- Rows = methods (baselines + ours), Columns = metrics/datasets
- Bold the best result in each column
- Include mean +/- std when available
- Use
\multirowfor method categories (Supervised, Self-supervised, etc.)
ablation — Ablation study table
- Rows = variants (full model, minus component A, minus component B, ...)
- Columns = metrics
- Bold the full model result
- Use checkmarks for component presence
descriptive — Dataset/statistics table
- Dataset characteristics, hyperparameters, or summary statistics
- Clean formatting with proper units
custom — Free-form table
- User specifies layout and content
Required Packages
\usepackage{booktabs} % \toprule, \midrule, \bottomrule
\usepackage{multirow} % \multirow
\usepackage{multicol} % multi-column layouts
\usepackage{threeparttable} % table notes
Output Format
Always generate tables with:
booktabsrules (\toprule,\midrule,\bottomrule)\caption{}and\label{tab:...}- Bold best results using
\textbf{} - Table notes via
threeparttablewhen needed - Proper alignment (
lfor text,corrfor numbers)
Rules
- Only include numbers from actual experimental logs — never hallucinate results
- All numbers must match the data source exactly
- Use
$\pm$for standard deviations - Use
\underline{}for second-best results when appropriate - Keep tables compact — avoid unnecessary columns
- Use
table*for wide tables spanning two columns - Add glossary/notes for abbreviated column headers
Related Skills
- Upstream: data-analysis, experiment-code
- Downstream: paper-writing-section, paper-compilation
- See also: figure-generation
Related skills
More from lingzhi227/agent-research-skills and the wider catalog.

algorithm-design
Design algorithms with LaTeX pseudocode and UML diagrams for formal method documentation.

atomic-decomposition
Decompose research papers into atomic concepts with bidirectional math-to-code mapping.

backward-traceability
Make every number in your PDF traceable to the exact code line that produced it.

citation-management
Manage BibTeX citations for LaTeX papers with Semantic Scholar integration and validation.

linkfox-multimodal-recognize-image
基于多模态AI的图片识别与分析。当用户想分析、描述、从图片URL中提取信息、image recognition, image analysis, image description, image content understanding, OCR text recognition, visual Q&A时触发此技能。当用户提到图片识别、图片分析、图片描述、识别图片内容、分析产品图、从图片中读取文字、描述图片、提取视觉内容或理解照片内容时触发。当用户提供图片URL并就其视觉内容提问时,即使未明确说"图片识别",也应触发此技能。

anti-render
Generate ideal-vs-reality visual comparisons by analyzing images and creating side-by-side contrasts.