StarVLA: A Lego-like Codebase for Vision-Language-Action Model Developing

[RSS 2026] Causal video-action world model for generalist robot control

Being-H is BeingBeyond's family of human-centric embodied foundation models.

InternVLA-A1: Unifying Understanding, Generation, and Action for Robotic Manipulation​

🎁 A collection of utilities for LeRobot.

[NeurIPS 2025] Official code for paper: Beyond Attention or Similarity: Maximizing Conditional Diversity for Token Pruning in MLLMs.

The repo of paper `RoboMamba: Multimodal State Space Model for Efficient Robot Reasoning and Manipulation`