[NeurIPS 2025] An official implementation of Flow-GRPO: Training Flow Matching Models via Online RL

Try OpenPI closed loop simulation / scenes generation / 3DGS model scenes in genie_sim

BFM_Zero: A Promptable Behavioral Foundation Model for Humanoid Control Using Unsupervised Reinforcement Learning

From Vision-Language-Action Models to a Real-World Robot Learning Stack

VLA-RFT: Vision-Language-Action Models with Reinforcement Fine-Tuning

Official repository for the project "TraceGen: World Modeling in 3D Trace-Space Enables Learning from Cross-Embodiment Videos" (CVPR'26)

[CoRL 2025] ManiFlow: A General Robot Manipulation Policy via Consistency Flow Training

Official code for "TraceGen: World Modeling in 3D Trace-Space Enables Learning from Cross-Embodiment Videos" (CVPR 2026)

Official Repository for RD-VLA

QDepth-VLA: Quantized Depth Prediction as Auxiliary Supervision for Vision-Language-Action Models

🔥 Official code repository for "Unlocking Dense Metric Depth Estimation in VLMs"

VL-JEPA Joint Embedding Predictive Architecture for Vision-language (Paper-Derived Implementation)Joint Embedding Predictive Architecture for Vision-language (Paper-Derived Implementation)

Official implementation of the ICML 2026 paper "DiLA: Disentangled Latent Action World Models".

XRZero-G0: Pushing the Frontier of Dexterous Robotic Manipulation with Interfaces, Quality and Ratios

[CVPR2026] Chain of World: World Model Thinking in Latent Motion

[RSS 2026] LDA-1B: Scaling Latent Dynamics Action Model via Universal Embodied Data Ingestion

This is a related study involving embodied intelligence based on modality.

VLA-JEPA: Enhancing Vision-Language-Action Model with Latent World Model

Scalable annotation pipeline for action-aglined fine-grained instruciton for Visual-language-Action model

NVIDIA Cosmos is an open platform of world models, datasets, and tools that enables developers to build Physical AI for robots, autonomous vehicles, smart infrastructure, and more.

PiStar: A closed-loop robot learning framework for data collection, value labeling, and policy fine-tuning across LIBERO simulation, Piper, and RealMan platforms.

[RSS 2026] Code for RISE: Self-Improving Robot Policy with Compositional World Model

☁️ Build multimodal AI applications with cloud-native stack

🎥 [Awesome] Egocentric / First-Person Video Datasets 📚 Papers, Benchmarks & Resources for Ego Vision

HumanEgo: Zero-Shot Robot Learning from Minutes of Human Egocentric Videos

openpi-RLT is an openpi-based real-robot RL system with RL-token-guided action refinement.

Tensor's VLA Training Infrastructure for Real-World Robotics in PyTorch