Hugging Face's logo

Datasets:
ArunKr
/
gui_grounding_dataset-1k

Tasks:
Text Generation
Modalities:
Text
Formats:
json
Languages:
English
Size:
1K - 10K
Tags:
agent
Libraries:
Datasets
pandas
License:
Dataset card Data Studio Files
xet
Community
Dataset Viewer
Auto-converted to Parquet
instruction
string
actions
list
Open Google homepage.
[ { "action": "open_url", "value": "https://google.com", "target": null, "options": null } ]
Upload file image.png.
[ { "action": "upload", "value": "notes.txt", "target": "input[type=file]", "options": null } ]
Check the terms box.
[ { "action": "check", "value": null, "target": "#terms", "options": null } ]
Type username Arun.
[ { "action": "type", "value": "TestUser", "target": "input#username", "options": null } ]
Accept popup.
[ { "action": "dialog", "value": "accept", "target": null, "options": null } ]
Search Google for Python tutorials.
[ { "action": "type", "value": "GUI grounding", "target": "textarea[name=q]", "options": null }, { "action": "keypress", "value": null, "target": null, "options": { "key": "Enter", "count": null, "direction": null } } ]
Accept popup.
[ { "action": "dialog", "value": "accept", "target": null, "options": null } ]
Check the terms box.
[ { "action": "check", "value": null, "target": "#terms", "options": null } ]
Take screenshot and save as image.png.
[ { "action": "screenshot", "value": "image.png", "target": null, "options": null } ]
Check the terms box.
[ { "action": "check", "value": null, "target": "#terms", "options": null } ]
Type username Bob.
[ { "action": "type", "value": "Bob", "target": "input#username", "options": null } ]
Open Google homepage.
[ { "action": "open_url", "value": "https://google.com", "target": null, "options": null } ]
Upload file image.png.
[ { "action": "upload", "value": "report.docx", "target": "input[type=file]", "options": null } ]
Check the terms box.
[ { "action": "check", "value": null, "target": "#terms", "options": null } ]
Check the terms box.
[ { "action": "check", "value": null, "target": "#terms", "options": null } ]
Take screenshot and save as resume.pdf.
[ { "action": "screenshot", "value": "notes.txt", "target": null, "options": null } ]
Accept popup.
[ { "action": "dialog", "value": "accept", "target": null, "options": null } ]
Type username Alice.
[ { "action": "type", "value": "Arun", "target": "input#username", "options": null } ]
Click the Submit button.
[ { "action": "click", "value": null, "target": "text=Next", "options": null } ]
Search Google for GUI grounding.
[ { "action": "type", "value": "Playwright automation", "target": "textarea[name=q]", "options": null }, { "action": "keypress", "value": null, "target": null, "options": { "key": "Enter", "count": null, "direction": null } } ]
Open Google homepage.
[ { "action": "open_url", "value": "https://google.com", "target": null, "options": null } ]
Type password qwerty.
[ { "action": "type", "value": "secret123", "target": "input[type=password]", "options": null } ]
Type password MyPass!.
[ { "action": "type", "value": "secret123", "target": "input[type=password]", "options": null } ]
Open Google homepage.
[ { "action": "open_url", "value": "https://google.com", "target": null, "options": null } ]
Search Google for AI.
[ { "action": "type", "value": "Playwright automation", "target": "textarea[name=q]", "options": null }, { "action": "keypress", "value": null, "target": null, "options": { "key": "Enter", "count": null, "direction": null } } ]
Type username TestUser.
[ { "action": "type", "value": "TestUser", "target": "input#username", "options": null } ]
Scroll down.
[ { "action": "scroll", "value": null, "target": null, "options": { "key": null, "count": 2, "direction": "{direction}" } } ]
Type username Admin.
[ { "action": "type", "value": "Admin", "target": "input#username", "options": null } ]
Search Google for Playwright automation.
[ { "action": "type", "value": "GUI grounding", "target": "textarea[name=q]", "options": null }, { "action": "keypress", "value": null, "target": null, "options": { "key": "Enter", "count": null, "direction": null } } ]
Upload file resume.pdf.
[ { "action": "upload", "value": "data.csv", "target": "input[type=file]", "options": null } ]
Check the terms box.
[ { "action": "check", "value": null, "target": "#terms", "options": null } ]
Click the Login button.
[ { "action": "click", "value": null, "target": "text=Submit", "options": null } ]
Scroll down.
[ { "action": "scroll", "value": null, "target": null, "options": { "key": null, "count": 2, "direction": "{direction}" } } ]
Search Google for AI.
[ { "action": "type", "value": "GUI grounding", "target": "textarea[name=q]", "options": null }, { "action": "keypress", "value": null, "target": null, "options": { "key": "Enter", "count": null, "direction": null } } ]
Search Google for GUI grounding.
[ { "action": "type", "value": "AI", "target": "textarea[name=q]", "options": null }, { "action": "keypress", "value": null, "target": null, "options": { "key": "Enter", "count": null, "direction": null } } ]
Check the terms box.
[ { "action": "check", "value": null, "target": "#terms", "options": null } ]
Accept popup.
[ { "action": "dialog", "value": "accept", "target": null, "options": null } ]
Check the terms box.
[ { "action": "check", "value": null, "target": "#terms", "options": null } ]
Check the terms box.
[ { "action": "check", "value": null, "target": "#terms", "options": null } ]
Upload file notes.txt.
[ { "action": "upload", "value": "resume.pdf", "target": "input[type=file]", "options": null } ]
Upload file data.csv.
[ { "action": "upload", "value": "resume.pdf", "target": "input[type=file]", "options": null } ]
Click the Search button.
[ { "action": "click", "value": null, "target": "text=Submit", "options": null } ]
Scroll up.
[ { "action": "scroll", "value": null, "target": null, "options": { "key": null, "count": 2, "direction": "{direction}" } } ]
Click the Confirm button.
[ { "action": "click", "value": null, "target": "text=Next", "options": null } ]
Upload file image.png.
[ { "action": "upload", "value": "resume.pdf", "target": "input[type=file]", "options": null } ]
Type username Arun.
[ { "action": "type", "value": "TestUser", "target": "input#username", "options": null } ]
Type password qwerty.
[ { "action": "type", "value": "qwerty", "target": "input[type=password]", "options": null } ]
Type password secret123.
[ { "action": "type", "value": "letmein", "target": "input[type=password]", "options": null } ]
Search Google for Playwright automation.
[ { "action": "type", "value": "GUI grounding", "target": "textarea[name=q]", "options": null }, { "action": "keypress", "value": null, "target": null, "options": { "key": "Enter", "count": null, "direction": null } } ]
Take screenshot and save as data.csv.
[ { "action": "screenshot", "value": "resume.pdf", "target": null, "options": null } ]
Click the Submit button.
[ { "action": "click", "value": null, "target": "text=Submit", "options": null } ]
Check the terms box.
[ { "action": "check", "value": null, "target": "#terms", "options": null } ]
Click the Submit button.
[ { "action": "click", "value": null, "target": "text=Login", "options": null } ]
Open Google homepage.
[ { "action": "open_url", "value": "https://google.com", "target": null, "options": null } ]
Accept popup.
[ { "action": "dialog", "value": "accept", "target": null, "options": null } ]
Check the terms box.
[ { "action": "check", "value": null, "target": "#terms", "options": null } ]
Check the terms box.
[ { "action": "check", "value": null, "target": "#terms", "options": null } ]
Upload file notes.txt.
[ { "action": "upload", "value": "report.docx", "target": "input[type=file]", "options": null } ]
Take screenshot and save as resume.pdf.
[ { "action": "screenshot", "value": "image.png", "target": null, "options": null } ]
Accept popup.
[ { "action": "dialog", "value": "accept", "target": null, "options": null } ]
Check the terms box.
[ { "action": "check", "value": null, "target": "#terms", "options": null } ]
Search Google for GUI grounding.
[ { "action": "type", "value": "machine learning", "target": "textarea[name=q]", "options": null }, { "action": "keypress", "value": null, "target": null, "options": { "key": "Enter", "count": null, "direction": null } } ]
Open Google homepage.
[ { "action": "open_url", "value": "https://google.com", "target": null, "options": null } ]
Search Google for machine learning.
[ { "action": "type", "value": "GUI grounding", "target": "textarea[name=q]", "options": null }, { "action": "keypress", "value": null, "target": null, "options": { "key": "Enter", "count": null, "direction": null } } ]
Take screenshot and save as notes.txt.
[ { "action": "screenshot", "value": "notes.txt", "target": null, "options": null } ]
Accept popup.
[ { "action": "dialog", "value": "accept", "target": null, "options": null } ]
Open Google homepage.
[ { "action": "open_url", "value": "https://google.com", "target": null, "options": null } ]
Scroll up.
[ { "action": "scroll", "value": null, "target": null, "options": { "key": null, "count": 2, "direction": "{direction}" } } ]
Type password letmein.
[ { "action": "type", "value": "MyPass!", "target": "input[type=password]", "options": null } ]
Click the Submit button.
[ { "action": "click", "value": null, "target": "text=Submit", "options": null } ]
Upload file notes.txt.
[ { "action": "upload", "value": "image.png", "target": "input[type=file]", "options": null } ]
Type username Alice.
[ { "action": "type", "value": "Bob", "target": "input#username", "options": null } ]
Search Google for Playwright automation.
[ { "action": "type", "value": "Python tutorials", "target": "textarea[name=q]", "options": null }, { "action": "keypress", "value": null, "target": null, "options": { "key": "Enter", "count": null, "direction": null } } ]
Take screenshot and save as notes.txt.
[ { "action": "screenshot", "value": "notes.txt", "target": null, "options": null } ]
Scroll down.
[ { "action": "scroll", "value": null, "target": null, "options": { "key": null, "count": 2, "direction": "{direction}" } } ]
Click the Confirm button.
[ { "action": "click", "value": null, "target": "text=Submit", "options": null } ]
Type password secret123.
[ { "action": "type", "value": "secret123", "target": "input[type=password]", "options": null } ]
Type username Admin.
[ { "action": "type", "value": "Bob", "target": "input#username", "options": null } ]
Type username TestUser.
[ { "action": "type", "value": "Bob", "target": "input#username", "options": null } ]
Take screenshot and save as report.docx.
[ { "action": "screenshot", "value": "notes.txt", "target": null, "options": null } ]
Search Google for AI.
[ { "action": "type", "value": "Python tutorials", "target": "textarea[name=q]", "options": null }, { "action": "keypress", "value": null, "target": null, "options": { "key": "Enter", "count": null, "direction": null } } ]
Take screenshot and save as resume.pdf.
[ { "action": "screenshot", "value": "image.png", "target": null, "options": null } ]
Open Google homepage.
[ { "action": "open_url", "value": "https://google.com", "target": null, "options": null } ]
Click the Submit button.
[ { "action": "click", "value": null, "target": "text=Confirm", "options": null } ]
Open Google homepage.
[ { "action": "open_url", "value": "https://google.com", "target": null, "options": null } ]
Click the Search button.
[ { "action": "click", "value": null, "target": "text=Login", "options": null } ]
Scroll up.
[ { "action": "scroll", "value": null, "target": null, "options": { "key": null, "count": 2, "direction": "{direction}" } } ]
Type password MyPass!.
[ { "action": "type", "value": "demo@123", "target": "input[type=password]", "options": null } ]
Scroll up.
[ { "action": "scroll", "value": null, "target": null, "options": { "key": null, "count": 2, "direction": "{direction}" } } ]
Open Google homepage.
[ { "action": "open_url", "value": "https://google.com", "target": null, "options": null } ]
Open Google homepage.
[ { "action": "open_url", "value": "https://google.com", "target": null, "options": null } ]
Open Google homepage.
[ { "action": "open_url", "value": "https://google.com", "target": null, "options": null } ]
Click the Confirm button.
[ { "action": "click", "value": null, "target": "text=Submit", "options": null } ]
Upload file report.docx.
[ { "action": "upload", "value": "report.docx", "target": "input[type=file]", "options": null } ]
Search Google for GUI grounding.
[ { "action": "type", "value": "Playwright automation", "target": "textarea[name=q]", "options": null }, { "action": "keypress", "value": null, "target": null, "options": { "key": "Enter", "count": null, "direction": null } } ]
Open Google homepage.
[ { "action": "open_url", "value": "https://google.com", "target": null, "options": null } ]
Type username TestUser.
[ { "action": "type", "value": "Alice", "target": "input#username", "options": null } ]
Take screenshot and save as image.png.
[ { "action": "screenshot", "value": "resume.pdf", "target": null, "options": null } ]
Accept popup.
[ { "action": "dialog", "value": "accept", "target": null, "options": null } ]
Search Google for Playwright automation.
[ { "action": "type", "value": "machine learning", "target": "textarea[name=q]", "options": null }, { "action": "keypress", "value": null, "target": null, "options": { "key": "Enter", "count": null, "direction": null } } ]
End of preview. Expand in Data Studio

Supported Tasks

  • Natural Language → GUI Action Grounding Convert user instructions into JSON action objects.
  • Instruction Following Models learn to interpret varying natural language formulations (e.g., “press submit” vs “click the submit button”).
  • Multi-step UI Automation Some samples involve sequences of actions (e.g., open site → type → press Enter → screenshot).

Languages

  • English (en)
  • Generated with simple variations (synonyms, phrasings).

Dataset Structure

Data Format

Each entry is a JSON object with:

{
  "instruction": "Search Google for Python Playwright.",
  "actions": [
    {"action": "type", "target": "textarea[name=q]", "value": "Python Playwright"},
    {"action": "keypress", "options": {"key": "Enter"}}
  ]
}
  • instruction: natural language input.
  • actions: list of structured GUI actions in a standard schema.

JSON Schema for Actions 

{
  "action": "string",
  "target": "string (CSS selector, text, XPath, etc.)",
  "value": "string (optional, e.g., input text, file path)",
  "options": { "key": "Enter", "button": "left", "count": 2, "direction": "down" }
}

Dataset Statistics

  • Size: 1,000 examples

  • Average actions per instruction: 1.7

  • Action types covered:

    • Click, double-click, right-click
    • Type, clear, keypress
    • Scroll, hover, drag-drop
    • Checkbox, radio, dropdown selection
    • Upload, download
    • Dialog handling (accept/dismiss)
    • Screenshot, highlight
    • Navigation (open_url, back, forward, refresh, switch_tab, resize)

Use Cases

  • Training GUI grounding agents (LLM-based or hybrid).
  • Creating instruction-tuned models for web automation.
  • Benchmarking natural language → structured action translation.
  • Bootstrapping RPA (Robotic Process Automation) agents with LLMs.

Generation Process

  • Synthetic data generated with templates + variations.
  • Actions derived from common web automation tasks (Google, YouTube, Gmail, Amazon, GitHub, Slack, Notion, Trello, etc.).
  • Covers both single-step (click, type) and multi-step (search + click + screenshot) workflows.

Limitations

  • Synthetic: No real human annotations.
  • Web-centric: Mostly web app actions, fewer desktop/native app actions.
  • Surface-level grounding: Uses simple selectors (text=Login, input#username) rather than pixel-perfect or accessibility trees.

Licensing

  • MIT License for dataset release.
  • Free for research & commercial use.
  • Attribution appreciated: “GUI Grounding Dataset (ArunKr, 2025)”.

Citation

If you use this dataset in your work, please cite:

@dataset{gui_grounding_2025,
  author    = {ArunKr},
  title     = {GUI Grounding Dataset},
  year      = {2025},
  url       = {https://huggingface.co/datasets/ArunKr/gui_grounding_dataset-1k},
  license   = {MIT}
}
Downloads last month
56
Size of downloaded dataset files:
122 kB
Size of the auto-converted Parquet files:
10.4 kB
Number of rows:
1,000