Div-Infinity/LISA
(NeurIPS '22) LISA: Learning Interpretable Skill Abstractions - A framework for unsupervised skill learning using Imitation
LISA helps create AI agents or robots that can understand and execute complex, language-based instructions in environments like navigation or robotic manipulation. It takes examples of tasks described in natural language and demonstrates how to complete them. The output is a set of 'skills' that the agent learns, allowing it to interpret new instructions and perform tasks more effectively. This is for researchers and engineers developing autonomous agents for complex, multi-step tasks.
No commits in the last 6 months.
Use this if you need to train AI agents or robots to perform sequential decision-making tasks based on language instructions, especially when you have limited training data.
Not ideal if your tasks are simple, don't require language interpretation, or if you prefer non-hierarchical learning approaches.
Stars
29
Forks
7
Language
Python
License
—
Category
Last pushed
Feb 22, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/nlp/Div-Infinity/LISA"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
Ayanami0730/deep_research_bench
DeepResearch Bench: A Comprehensive Benchmark for Deep Research Agents
Hsankesara/DeepResearch
This repository is the collection of research papers in Deep learning, computer vision and NLP.
QizhiPei/Awesome-Biomolecule-Language-Cross-Modeling
Awesome-Biomolecule-Language-Cross-Modeling: a curated list of resources for paper "Leveraging...
thuiar/OKD-Reading-List
Papers for Open Knowledge Discovery
roomylee/nlp-papers-with-arxiv
Statistics and accepted paper list of NLP conferences with arXiv link