George121380/LEAP

A framework for continual behavior learning in embodied agents through interaction and human guidance.

28
/ 100
Experimental

This project helps researchers and engineers design and train intelligent robots or virtual agents that can learn new tasks over time, mimicking how humans learn through experience and instruction. You provide natural language instructions and an interactive environment, and the system outputs an agent capable of performing complex tasks in a household setting, continually improving its behavior. It's ideal for those developing advanced AI for robotics, virtual assistants, or simulation environments.

No commits in the last 6 months.

Use this if you need an agent to learn complex, multi-step tasks in an interactive environment, adapting to new instructions and accumulating knowledge over its 'lifetime' without being reset.

Not ideal if you're looking for a pre-trained, ready-to-deploy agent for a specific, narrow task or if your environment is purely static without opportunities for interaction or human feedback.

robotics-learning embodied-AI human-robot-interaction lifelong-learning virtual-agent-training
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 5 / 25
Maturity 15 / 25
Community 6 / 25

How are scores calculated?

Stars

12

Forks

1

Language

Python

License

MIT

Last pushed

Aug 17, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/George121380/LEAP"

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