# LIV: Language-Image Representations and Rewards for Robotics Control
International Conference on Machine Learning (ICML), 2023

[[Project Page]](https://penn-pal-lab.github.io/LIV) [[Arxiv]](https://arxiv.org/abs/2306.00958) [[Paper]](https://www.seas.upenn.edu/~jasonyma/liv_paper.pdf) [[Dataset]](liv/dataset/README.md) 

[Jason Yecheng Ma](https://www.seas.upenn.edu/~jasonyma/)<sup>1</sup>, [Vikash Kumar](https://vikashplus.github.io/)<sup>2</sup>, [Amy Zhang](https://amyzhang.github.io/)<sup>2</sup>, [Osbert Bastani](https://obastani.github.io/)<sup>1</sup>, [Dinesh Jayaraman](https://www.seas.upenn.edu/~dineshj/)<sup>1</sup>

<sup>1</sup>University of Pennsylvania, <sup>2</sup>Meta AI

This repository hosts pre-trained LIV model released in our ICML paper.