zjunlp/WorldMind

Aligning Agentic World Models via Knowledgeable Experience Learning

39
/ 100
Emerging

This project helps researchers and developers working with embodied AI agents create systems that learn and adapt to new tasks and environments more effectively. It takes in an agent's interactions within a simulated environment, like a robot navigating a house or rearranging objects, and outputs reusable knowledge and causal rules. This knowledge enables agents to generalize behaviors without extensive retraining. It's designed for AI researchers and practitioners building sophisticated embodied agents.

Use this if you are developing embodied AI agents and want them to learn from experience, generalize across different tasks (like household chores or object rearrangement), and continuously improve their understanding of the environment without constant retraining.

Not ideal if you are working with AI agents that operate in purely abstract, non-physical environments or if your agents do not need to learn and adapt to dynamic, interactive surroundings.

Embodied AI Robotics Agent Learning Reinforcement Learning Human-Robot Interaction
No Package No Dependents
Maintenance 10 / 25
Adoption 7 / 25
Maturity 13 / 25
Community 9 / 25

How are scores calculated?

Stars

31

Forks

3

Language

Python

License

MIT

Last pushed

Jan 25, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/agents/zjunlp/WorldMind"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.