Vision-grounded Integration for Reward design And Learning is inspired by DREFUN-V. This project aims to create a framework for reward function generation in Gymnasium environments, utilizing and VideoLLM in order to enchance aligment too. Below, you can see how the framework is structured.
For more details about the diagram, you can refer to our paper: paper
You can visit our website, which provides installation instructions as well as project documentation:
We also provide a link to the course by Mr. YUN that we attended: