Dataset Viewer
Auto-converted to Parquet Duplicate
video_id
string
step
int32
system
string
user
string
assistant
string
image
image
video_001
0
You are an AI assistant analyzing high-level user tasks from screen recordings.
Analyze the user's high-level task based on the 'before' and 'after' screenshots. Task to Perform: Writing/Replying Emails
<think> This frame pair shows the execution of the atomic task: 'Writing/Replying Emails'. The task starts at timestamp 88s (before) and is completed by timestamp 118s (after). The corresponding high-level action is captured in the code block. </think> <code> send_email(to="khawajamuraddulhassan@gmail.com", subject="AI...
video_001
1
You are an AI assistant analyzing high-level user tasks from screen recordings.
Analyze the user's high-level task based on the 'before' and 'after' screenshots. Task to Perform: Posting on linkedin
<think> This frame pair shows the execution of the atomic task: 'Posting on linkedin'. The task starts at timestamp 17s (before) and is completed by timestamp 23s (after). The corresponding high-level action is captured in the code block. </think> <code> post_on_linkedin(content="Exciting Opportunity at Dummy Company!�...

UI Automation Dataset (Multi-Video)

2 examples from 1 videos - UI automation tasks from screen recordings.

Dataset Structure

Each entry contains:

  • video_id: Sequential ID for each video (video_001, video_002, etc.)
  • step: Step number within that video (0, 1, 2, ...)
  • system: System prompt for the GUI agent
  • user: Task instruction + previous actions
  • assistant: Model's reasoning and action
  • image: Screenshot of the UI state

Usage

from datasets import load_dataset

ds = load_dataset("KMH158-QLU/recruiter_atomic_task_final7")

# Access by video
for video_id in set(ds['train']['video_id']):
    video_data = ds['train'].filter(lambda x: x['video_id'] == video_id)
    print(f"Video {video_id}: {len(video_data)} steps")

# Or iterate all examples
for item in ds['train']:
    print(f"{item['video_id']} - Step {item['step']}: {item['assistant'][:50]}...")

Growing Dataset

This dataset supports multiple videos. Each video gets a unique ID (video_001, video_002, etc.). New videos are automatically appended with the next available ID.

Downloads last month
5