Question
stringlengths 17
231
| Category
stringclasses 5
values |
|---|---|
Add a column with the yearly growth in gdp per country
|
Classic Data Analysis
|
Add a column with the capital of each country
|
Classic Data Analysis
|
Add a column that checks if the capital of the country is above or below the equator
|
Classic Data Analysis
|
Classify the countries in “Developed” if >$30k, else “Emerging.”
|
Classic Data Analysis
|
Add a table with the approximate population of each country
|
Classic Data Analysis
|
Add a column with the name of the currency used by each country and the exchange rate to USD.
|
Classic Data Analysis
|
Extract the ISO 3166-1 alpha-2 from each country
|
Classic Data Analysis
|
Which country had the fastest growth in GDP per capita (CAGR) between 2000 and 2020?
|
Classic Data Analysis
|
Which country overtook the US GDP per capita by 2020?
|
Classic Data Analysis
|
Did China overtake Brazil in GDP per capita? If so, when?
|
Classic Data Analysis
|
Which G7 country had the slowest growth 2000–2020?
|
Classic Data Analysis
|
Identify the first year China exceeded $10,000 GDP per capita.
|
Classic Data Analysis
|
What was the GDP per capita for Germany in 2010.
|
Classic Data Analysis
|
Which African country grew the fastest since 2000?
|
Classic Data Analysis
|
Plot GDP per capita trend for USA, China, India, Brazil.
|
Classic Data Analysis
|
Create a bar chart of top 10 GDP per capita countries in 2020.
|
Classic Data Analysis
|
Change the bar chart color to green
|
Classic Data Analysis
|
Change the title of the chart to Top 10 GDP per capita summary
|
Classic Data Analysis
|
Create a scatterplot of GDP per capita vs growth rate (2020).
|
Classic Data Analysis
|
Produce a heatmap of GDP per capita growth rates by country-year.
|
Classic Data Analysis
|
Annotate recession periods in US GDP per capita.
|
Classic Data Analysis
|
Suggest best chart type to compare top 10 in 2000 vs 2020.
|
Classic Data Analysis
|
Merge the dataset with the capitals table
|
Classic Data Analysis
|
Check the Capitals table for inconsistencies and data issues.
|
Classic Data Analysis
|
Remove duplicates from the Capitals table
|
Classic Data Analysis
|
Use K-Means to cluster countries in 5 clusters and walk me through what they mean
|
Advanced Analysis
|
Group countries into quartiles based on GDP per capita.
|
Classic Data Analysis
|
Do a matrix with the years before/after joining EU (-5 to +10) and the yoy gdp per capita growth. Consider only the 2004 and 2007 & 2013 (joined) cohorts
|
Classic Data Analysis
|
Write the formula to calculate CAGR for GDP per capita for Germany (2000–2020).
|
Classic Data Analysis
|
Summarize the growth in GDP per capita in the last 10 years for OECD countries in a table.
|
Classic Data Analysis
|
Rank countries by GDP per capita in 2020.
|
Classic Data Analysis
|
Compare OECD vs. non-OECD GDP per capita trends since 1990. Don't ask follow up questions. Assume whatever you find are the best options.
|
Classic Data Analysis
|
Identify countries with ≥20% GDP per capita drop in a single year.
|
Classic Data Analysis
|
Which country showed sharp GDP per capita drop in 2020 but full recovery in 2021?
|
Classic Data Analysis
|
Compare Eastern Europe vs. Western Europe convergence since 1990. Don't ask follow up questions. Assume whatever you find are the best options.
|
Classic Data Analysis
|
Calculate 5-year moving average for India.
|
Classic Data Analysis
|
Create a dashboard that shows the evolution of GDP per capita over time. Split by region, show best and worst country performs and summarize the results. Don't ask follow up questions. Assume whatever you find are the best options.
|
Classic Data Analysis
|
Analyze if the countries that joined the EU after 2003 are converging or diverging from EU GDP. Don't ask follow up questions. Assume whatever you find are the best options.
|
Classic Data Analysis
|
Predict the growth in GPD per year for the next 5 years for each BRIC country.
|
Advanced Analysis
|
Flag anomalies in oil-rich economies linked to oil price crashes.
|
Advanced Analysis
|
Create a scatter plot of GDP per capita vs. population, with bubble size proportional to country area and color by continent.
|
Advanced Analysis
|
Make a boxplot of of the gdp per capita in 2023 per continent.
|
Advanced Analysis
|
Create a country head-off tool to compare the GPD per capita and growth between two countries since 1990
|
Creating models
|
Conditional: Highlight countries < $1,000 GDP per capita in 2020.
|
Manage Spreadsheet Elements
|
Sort the countries by alphabetical order of their capital
|
Manage Spreadsheet Elements
|
Freeze the panes in the table, format GDP as $
|
Manage Spreadsheet Elements
|
what is 8.9-8.11?
|
Arithmetic operations
|
Create a simulator to compare two investment strategies with equal monthly deposits over the same years but different annual rates. Show total returns, their difference, and a yearly return chart.
|
Creating models
|
what is 12.03 – 12.009?
|
Arithmetic operations
|
what is 7.100 + 7.01?
|
Arithmetic operations
|
what is 0.009 – 0.0011?
|
Arithmetic operations
|
what is 45.0002 ÷ 0.002?
|
Arithmetic operations
|
what is 3.0005 × 1.05?
|
Arithmetic operations
|
YAML Metadata
Warning:
empty or missing yaml metadata in repo card
(https://huggingface.co/docs/hub/datasets-cards)
AI Spreadsheet Benchmark Dataset
The AI Spreadsheet Benchmark captures 53 realistic spreadsheet prompts spanning analysis, enrichment, visualization, and workbook-management workflows. It is designed to evaluate how spreadsheet copilots behave in situ: Do they write formulas? Do charts stay linked to data? Can the output recompute when numbers change?
- Paper: "The AI Spreadsheet Benchmark: Measuring Dynamic Output in Spreadsheet Assistants"
- Dataset:
rowshq/aispreadsheetbenchmark - Metrics: Pass@1, Pass@3, Dynamic Output Rate, Latency
Accessing the dataset
from datasets import load_dataset
benchmark = load_dataset("rowshq/aispreadsheetbenchmark")
questions = benchmark["questions"] # prompt text, categories, scoring metadata
Task categories
| Category | Tasks | Examples |
|---|---|---|
| Classic Data Analysis | 37 | YoY growth columns, lookups, joins, dashboards, cohort tables |
| Advanced Analysis | 5 | K-means clustering, forecasting, anomaly detection, custom visualizations |
| Creating Models | 2 | Interactive head-to-head calculators, investment simulators |
| Manage Spreadsheet Elements | 3 | Conditional formatting, sorting, chart styling, sheet setup |
| Arithmetic Operations | 6 | High-precision arithmetic sanity checks |
Each prompt record contains:
- Natural-language instructions
- Category and sub-category labels
Recommended evaluation protocol
- Reset the workbook to the canonical dataset before each run.
- Issue prompts verbatim. If the assistant asks for clarification, respond neutrally while keeping the task scope fixed.
- Assess success using the published acceptance criteria (execution checks whenever possible).
- Assess dynamic output by perturbing underlying data and verifying the response updates automatically (no pasted values or screenshots).
- Measure latency from prompt submission to assistant completion.
- Compute metrics: Pass@1, Pass@3 (up to three attempts per task), Dynamic Output Rate, mean/median latency.
Baseline results
Initial evaluation across five assistants (Rows AI Analyst, Excel Copilot, Google Sheets + Gemini, Shortcut, Julius):
| Assistant | Pass@1 | Pass@3 | Dynamic (%) | Mean time (s) |
|---|---|---|---|---|
| Rows AI Analyst | 89 | 92 | 74 | 220 |
| Microsoft Excel Copilot | 53 | 64 | 8 | 46 |
| Google Sheets + Gemini | 57 | 64 | 6 | 11 |
| Shortcut | 83 | 83 | 13 | 222 |
| Julius | 75 | 83 | 0 | 30 |
Detailed per-category tables and visualizations appear in the accompanying technical paper.
Citation
@misc{rowshq2025benchmark,
title = {The AI Spreadsheet Benchmark: Measuring Intelligence in Spreadsheet Assistants},
author = {Samagaio, Álvaro Mendes and Cruz, Henrique and Pereira, Humberto Ayres and Schulz, Torben},
year = {2025},
url = {https://huggingface.co/datasets/rowshq/aispreadsheetbenchmark/blob/main/technical_paper.pdf}
}
Questions & contributions
Open a discussion or issue on the Hugging Face dataset page if you:
- Find discrepancies in acceptance criteria or scoring instructions
- Want to share new assistant baselines or evaluation tooling
- Plan to extend the benchmark with additional domains or datasets
We welcome contributions that improve documentation, acceptance criteria, or reproducibility assets. Reach out via the dataset page to coordinate substantial updates.
- Downloads last month
- 122