Open-source alternatives to Grape

Interactive Post-Training for Vision-Language-Action Models Official implementation of RIPT-VLA. Parts of the repo are built on a fork of QueST.

We provide examples to fine-tune Octo, on the top of HIL-SERL that provides the base environment to perform robotic manipulation tasks with human interventions. The following sections describe how to use our code.

This repo contains code for the paper:

This repository contains the code for the paper What Can RL Bring to VLA Generalization? An Empirical Study. The pretrained checkpoints are available at HuggingFace.

🎯 General Manipulation: Improving OpenVLA-7B with outcome-based multi-task reinforcement learing.

📃 Main Results • ✨ Getting Started 🌻 Acknowledgement • 📨 Contact • 📝 TODO • 🎈 Citation --> SimpleVLA-RL is an efficient RL framework for VLA that improves long-horizon planning under data scarcity. It leverages reinforcement learning that can substantially outperforms SFT in simulation and…