Synthux Economy R8 W101: Simulated Desktop Computer-Use Trajectories

Grounded visual computer-use trajectories generated by the SynthUX system. Each record captures one worker's device session inside a simulated company, where a goal expands into a node tree and terminal nodes drive real desktop-simulator applications through low-level mouse and keyboard input. The observed trajectory is recorded as a screen video plus per-node frames, as described by author jacob-valdez on Hugging Face.

Use Cases

• Training or benchmarking AI agents for desktop task automation based on low-level mouse and keyboard input sequences.
• Studying human-computer interaction patterns based on recorded screen videos and per-node frames of task execution.
• Developing models for goal decomposition and task planning based on the described node-tree structure of user goals.
• Analyzing multi-application workflow efficiency based on trajectories spanning Terminal, Notes, VS Code, Browser, Slack/Teams, Mail, Sheets, and Slides.

Strengths

• Data includes screen video recordings, providing a rich visual record of user interactions.
• Trajectories are grounded in a simulated company environment with specific user goals, suggesting contextual relevance.
• The description mentions a node-tree structure for goal expansion, indicating a structured representation of tasks.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment for large-scale model training.
• The dataset's synthetic nature may not fully capture the complexity or biases of real-world user behavior.

Provenance

Source

Hugging Face dataset uploaded by jacob-valdez.

Collection Method

Generated by the SynthUX system using a simulated desktop environment.

Freshness

Last updated 2026-06-02 03:00:44; freshness should be verified.