End-to-end pipeline converting generative videos (Veo, Sora) to humanoid robot motions

LIBERO-PRO is the official repository of the LIBERO-PRO — an evaluation extension of the original LIBERO benchmark

Grounded SAM 2: Ground and Track Anything in Videos with Grounding DINO, Florence-2 and SAM 2

Code for "Novel Object 6D Pose Estimation with a Single Reference View".

Fast-in-Slow: A Dual-System Foundation Model Unifying Fast Manipulation within Slow Reasoning

6D Cartesian space hybrid force-velocity control using positional inner loop and wrist mounted FT sensor.

Official implementation for Compliant Residual DAgger

🔥 SpatialVLA: a spatial-enhanced vision-language-action model that is trained on 1.1 Million real robot episodes. Accepted at RSS 2025.

a Lightweight Motion Planning Package

WoW (World-Omniscient World Model) is a generative world model trained on 2 million robotic interaction trajectories, designed to imagine, reason, and act in the physical world. Unlike passive vide…

Code for PEEK: Guiding and Minimal Image Representations for Zero-Shot Generalization of Robot Manipulation Policies

Official implementation of ReconVLA: Reconstructive Vision-Language-Action Model as Effective Robot Perceiver.

VLA-Adapter: An Effective Paradigm for Tiny-Scale Vision-Language-Action Model

[AAAI 2026] Official code for MoLe-VLA: Dynamic Layer-skipping Vision Language Action Model via Mixture-of-Layers for Efficient Robot Manipulation

Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success

Repo for running various baselines with Behavior-1K

Team Comet's 2025 BEHAVIOR Challenge Codebase

This is a official code for the benchmark of the paper "VTDexManip: A Dataset and Benchmark for Visual-tactile Pretraining and Dexterous Manipulation with Reinforcement Learning" (ICLR 2025)

Extract frames and motion vectors from H.264 and MPEG-4 encoded video.

HiF-VLA: An efficient, bidirectional spatiotemporal expansion Vision-Language-Action Model

egocentric humanoid manipulation benchmark

[ICLR 25] Code for "Efficient Diffusion Transformer Policies with Mixture of Expert Denoisers for Multitask Learning"

Uni-Hand: Universal Hand Motion Forecasting in Egocentric Views (with visual imitation learning for robots)

Learning Dexterous Manipulation Skills from Imperfect Simulations

Finetuning Offline World Models in the Real World

Official Implementation of "Real-world RL for Active Perception Behaviors"

[CVPR 2025]Lift3D Foundation Policy: Lifting 2D Large-Scale Pretrained Models for Robust 3D Robotic Manipulation

Official Release of "Mixture of Horizons in Action Chunking"

MM-ACT: Learn from Multimodal Parallel Generation to Act