HEX is a whole-body vision-language-action framework for full-sized humanoid robots.
LAP: Language-Action Pre-Training Enables Zero-Shot Cross Embodiment Transfer
LocoFormer - Generalist Locomotion via Long-Context Adaptation
Official implementation of the ICML 2026 paper "DiLA: Disentangled Latent Action World Models".
Awesome robot data engines for VLA: collection, synthesis, augmentation, curation, preprocessing, and benchmarks.
Repository hosting the official code of the paper "PCHands: PCA-based Hand Pose Retargeting on Manipulators with N-DoF"
🤖 Explore LocoFormer, a Transformer-XL model that enhances robot locomotion through long-context learning and real-world adaptability.
Curated papers, datasets, systems, and benchmarks for robot data engines across robot-centric, UMI, human/egocentric, and simulation data.
Documentation for a frozen force-control skill that ports across position/velocity-commanded cobots via a locked interface contract. Validated on Kinova Gen3.
Paper/Code list of cross-emboided.
X-Embodiment Language-Grounded Manipulation Benchmark: measuring language-conditioned policy transfer between a Panda arm and a Unitree G1 humanoid in ManiSkill3
Answers whether several robot demonstration datasets can be mixed by inspecting their manifests alone, scoring mixability from action space, control frequency, gripper convention, and metadata gaps.
A versioned, checkable package format for robot skills: strict schema, content-hash provenance and payload integrity, a declared cross-embodiment adaptation contract, safety verification against robot limits and preconditions, and validated composition. Reference CLI and library, no robot needed.
Human mocap retargeting and whole-body motion tracking for humanoids (Unitree G1, IGRIS-C) in Isaac Lab and MuJoCo. Simulation only.
To associate your repository with the cross-embodiment topic, visit your repo's landing page and select "manage topics."