DENG HAOYUAN

SoftVTBench. visuotactile benchmark for the deformable object manipulation

Dream-Tac. Official Code for Dream-Tac

UniHand. [TPAMI 2026] Uni-Hand: Universal Hand Motion Forecasting in Egocentric Views (with Visual Imitation Learning for Robots)

LIBERO-plus. Official repository of LIBERO-plus, a generalized benchmark for in-depth robustness analysis of vision-language-action models.

N0-Foundation. 𝒩0-Foundation: Towards the Age of Tactile Intelligence

N0-TWAM. N₀-TWAM: A Tactile-Native World Action Model for Contact-Rich Manipulation

N0-VTLA. N0-VTLA: Scaling Vision-Tactile-Language-Action Model with Latent Tactile Tokens

E2HiL-project-a1x. Python

Hy-Embodied-0.5-VLA. From Vision-Language-Action Models to a Real-World Robot Learning Stack

Xiaomi-Robotics-1. Code for Xiaomi-Robotics-1

rlt-openpi. Open-source implementation of "RL Token: Bootstrapping Online RL with Vision-Language-Action Models"

StarWAM. A Generalizable Codebase for World-Action Models

HY-World-2.0. HY-World 2.0: A Multi-Modal World Model for Reconstructing, Generating, and Simulating 3D Worlds

DiffusionNFT. [ICLR 2026 Oral] DiffusionNFT: Online Diffusion Reinforcement with Forward Process

articraft. Superseded by https://github.com/articraftresearch/Articraft. This repository is kept for reference.

newton. An open-source, GPU-accelerated physics simulation engine built upon NVIDIA Warp, specifically targeting roboticists and simulation researchers.

DiT4DiT. This is the official code repo for DiT4DiT, a Vision-Action-Model (VAM) framework that combines video generation model with flow-matching-based action prediction for generalizable robotic manipulation.

iMaC. iMac: Translating Actions into Motion and Contact Images for Embodied World Models

Awesome-Progress-Reward-Model.

Awesome-Video-Action-Model.

RL-100. Implementation of RL-100, Performant Robotic Manipulation with Real-World Reinforcement Learning

Awesome-WAM. A curated, continuously updated reading list, paper blogs, and resources for World Action Models (WAMs) in embodied AI.

VLA-JEPA. [ECCV 2026] VLA-JEPA: Enhancing Vision-Language-Action Model with Latent World Model

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

Awesome-World-Models. A comprehensive list of papers for the definition of World Models and using World Models for General Video Generation, Embodied AI, and Autonomous Driving, including papers, codes, and related websites.

faster. Python

vla_foundry. Python

awesome-embodied-vla-va-vln. A curated list of state-of-the-art research in embodied AI, focusing on vision-language-action (VLA) models, vision-language navigation (VLN), and related multimodal learning approaches.

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

cap-x. A Framework for Benchmarking and Improving Coding Agents for Robot Manipulation

FastWAM. Official codebase for Fast-WAM: Do World Action Models Need Test-time Future Imagination?

ABot-PhysWorld. Python

Awesome-RL-for-LRMs. A Survey of Reinforcement Learning for Large Reasoning Models

Embodied-AI-Paper-TopConf. [Actively Maintained🔥] A list of Embodied AI papers accepted by top conferences (ICLR, NeurIPS, ICML, RSS, CoRL, ICRA, IROS, CVPR, ICCV, ECCV).

Awesome-RL-for-Video-Generation. A curated list of papers on reinforcement learning for video generation

dsrl_pi0. Official implementation for pi0 steering via DSRL, Steering Your Diffusion Policy with Latent Space Reinforcement Learning (CoRL 2025)

ULIP. Python

robometer. Robometer: Scaling General-Purpose Robotic Reward Models via Trajectory Comparisons

Genie-Envisioner-V1. Python

robocasa. RoboCasa: Large-Scale Simulation of Everyday Tasks for Generalist Robots

Evo-RL. We release Evo-RL, the opensource real-world offline RL on So-101 and AgileX PiPER for easier reproduction.

VLAExplain. VLA model interpretability tools

pi-StepNFT. Python

dev-browser. A Claude Skill to give your agent the ability to use a web browser

cover-vla. This is the official codebase for paper: Scaling Verification Can Be More Effective than Scaling Policy Learning for Vision-Language-Action Alignment

agentskills. Specification and documentation for Agent Skills

miles. Miles is an enterprise-facing reinforcement learning framework for LLM and VLM post-training, forked from and co-evolving with slime.

microgpt-rust. An atomic way to train and inference a GPT in Rust and Go. Inspired by @karpathy.

AnyTouch2. [ICLR 2026] AnyTouch 2: General Optical Tactile Representation Learning For Dynamic Tactile Perception

code. VLS: Steering Pretrained Robot Policies via Vision–Language Models

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

world-gymnast. World-Gymnast: Training Robots with Reinforcement Learning in a World Model

GreenVLA. Green-VLA: Staged Vision-Language-Action Model for Generalist Robots

dmpo-release. DMPO: Diffusion Model Policy Optimization

vla-scratch. Python

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

Awesome-Touch. Tactile Sensing • Data Collection • IL/RL/VLA/WM • Manipulation • Simulation • Open Source

DiffTactile. [ICLR 2024] DiffTactile: A Physics-based Differentiable Tactile Simulator for Contact-rich Robotic Manipulation

agent-lightning. The absolute trainer to light up AI agents.

vlagents. VLA policy server for simulation evaluation and real world rollout/deployment

every-embodied. 仅需Python基础，从0构建自己的具身智能机器人；从0逐步构建VLA/OpenVLA/SmolVLA/Pi0， 深入理解具身智能

bridger. Python

octopi. Python

VLA-Touch. Implementation of RA-L (2026) paper: VLA-Touch: Enhancing Vision-Language-Action Models with Dual-Level Tactile Feedback

DeepTutor. DeepTutor: Lifelong Personalized Tutoring. https://deeptutor.info/.

Awesome-Force-Tactile-VLA. A paper list of multimodal VLAs

Dream-VLX. Dream-VL and Dream-VLA, a diffusion VLM and a diffusion VLA.

RoboCOIN. RoboCoin + LeRobot integration

VisualGRPO. E-GRPO: High Entropy Steps Drive Effective Reinforcement Learning for Flow Models

learning_thousand_tasks. This repository contains the implementation of all methods evaluated in the paper "Learning a Thousand Tasks in a Day". We provide model architectures, training scripts, and deployment examples.

World-In-Your-Hands. Python

TwinAligner. [arxiv 2025] TwinAligner: Visual-Dynamic Alignment Empowers Physics-aware Real2Sim2Real for Robotic Manipulation

Real2Edit2Real. [CVPR 2026] Real2Edit2Real: Generating Robotic Demonstrations via a 3D Control Interface

REALM. [IEEE RA-L 2026] REALM: A Real-to-Sim Validated Benchmark for Generalization in Robotic Manipulation

Xbotics-Embodied-Guide. Xbotics 社区具身智能学习指南：我们把“具身综述→学习路线→仿真学习→开源实物→人物访谈→公司图谱”串起来，帮助新手和实战者快速定位路径、落地项目与参与开源。

MIND-V. Python

agentlace. Connect agent policies for distributed ML applications

MM-ACT. [CVPR'2026] "MM-ACT: Learn from Multimodal Parallel Generation to Act"

Awesome-World-Models. A Curated List of Awesome Works in World Modeling, Aiming to Serve as a One-stop Resource for Researchers, Practitioners, and Enthusiasts Interested in World Modeling.

trl. Train transformer language models with reinforcement learning.

MiMo-Embodied. MiMo-Embodied

Depth-Anything-3. Depth Anything 3

siiRL. siiRL: Shanghai Innovation Institute RL Framework for Advanced LLMs and Multi-Agent Systems

sam3. The repository provides code for running inference and finetuning with the Meta Segment Anything Model 3 (SAM 3), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.

sam-3d-objects. SAM 3D Objects

inverse-mixed-strategy. (ICRA 2025) Inverse Mixed Strategy Games with Generative Trajectory Models

DeepThinkVLA. DeepThinkVLA: Enhancing Reasoning Capability of Vision-Language-Action Models

nora-1.5. NORA-1.5: A Vision-Language-Action Model Trained using World Model- and Action-based Preference Rewards

G0. Galaxea's VLA

hiors. Human-in-the-loop Online Rejection Sampling for Robotic Manipulation

MoMaGen. Python

X-VLA. [ICLR 2026] The offical Implementation of "Soft-Prompted Transformer as Scalable Cross-Embodiment Vision-Language-Action Model"

Cal-QL. official implementation for our paper Cal-QL: Calibrated Offline RL Pre-Training for Efficient Online Fine-Tuning (NeurIPS 2023)

Awesome-RL-VLA. A Survey on Reinforcement Learning of Vision-Language-Action Models for Robotic Manipulation

Ctrl-World. ICLR 2026 Paper: Ctrl-World

SimPublisher. A cross-environment tool to publish objects from simulation for Augmented Reality and Human Robot Interaction.

Taccel. Taccel: Scaling-up Vision-based Tactile Robotics with High-performance GPU Simulation

RDT2. Official code of RDT 2

SoM. [arXiv 2023] Set-of-Mark Prompting for GPT-4V and LMMs

dppo. Official implementation of Diffusion Policy Policy Optimization, arxiv 2024

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

VLAC. VLAC: A Vision-Language-Action-Critic Model for Robotic Real-World Reinforcement Learning

DriveMoE. [CVPR 2026] Drive-π0 and DriveMoE on End-to-end Autonomous Driving

vjepa2. PyTorch code and models for VJEPA2 self-supervised learning from video.

LiveCodeBench. Official repository for the paper "LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code"

lerobot-sim2real. LeRobot sim2real code. Train in fast simulation and deploy visual policies zero shot to the real world

BEHAVIOR-1K. BEHAVIOR-1K: a platform for accelerating Embodied AI research. Join our Discord for support: https://discord.gg/bccR5vGFEx

detectron2. Detectron2 is a platform for object detection, segmentation and other visual recognition tasks.

GalaxeaDP. Galaxea's first diffusion policy release

RLinf. RLinf: Reinforcement Learning Infrastructure for Embodied and Agentic AI

InternManip. An All-in-one robot manipulation learning suite for policy models training and evaluation on various datasets and benchmarks.

Dex1B. Implementation of Dex1B: Learning with 1B Demonstrations for Dexterous Manipulation, from Ye et al of UCSD

H_RDT. H-RDT: Human Manipulation Enhanced Bimanual Robotic Manipulation

flowmatching_policy_rl. Python

qc. Python

EmbodieDreamer.

Tool_as_Interface. Python

mast3r. Grounding Image Matching in 3D with MASt3R

arp. Autoregressive Policy for Robot Learning (RA-L 2025)

E2FGVI. Official code for "Towards An End-to-End Framework for Flow-Guided Video Inpainting" (CVPR2022)

rovi-aug. Augment robotics demonstration datasets with different robots and viewpoints

Entropy-Mechanism-of-RL. The Entropy Mechanism of Reinforcement Learning for Large Language Model Reasoning.

roboengine. Official Reporsitory of "RoboEngine: Plug-and-Play Robot Data Augmentation with Semantic Robot Segmentation and Background Generation"

dex-retargeting. Various retargeting optimizers to translate human hand motion to robot hand motion.

ViTaL. Accompanying codebase for paper"Touch begins where vision ends: Generalizable policies for contact-rich manipulation"

Awesome-Image-Harmonization. A curated list of papers, code and resources pertaining to image harmonization.

COCOCO. Video-Inpaint-Anything: This is the inference code for our paper CoCoCo: Improving Text-Guided Video Inpainting for Better Consistency, Controllability and Compatibility.

robot-colosseum. A Benchmark for Evaluating Generalization for Robotic Manipulation

YOTO. [TPAMI2026 / RSS2025] Code for my paper "You Only Teach Once: Learn One-Shot Bimanual Robotic Manipulation from Video Demonstrations"

Inpaint-Anything. Inpaint anything using Segment Anything and inpainting models.

Video-Depth-Anything. [CVPR 2025 Highlight] Video Depth Anything: Consistent Depth Estimation for Super-Long Videos

nora. NORA: A Small Open-Sourced Generalist Vision Language Action Model for Embodied Tasks

MinerU. Transforms complex documents like PDFs and Office docs into LLM-ready markdown/JSON for your Agentic workflows.

ManiGaussian_Bimanual. [IROS 2025] ManiGaussian++: General Robotic Bimanual Manipulation with Hierarchical Gaussian World Model

Isaac-GR00T. NVIDIA Isaac GR00T N1.7 - A Foundation Model for Generalist Robots.

Embodied-AI-Guide. [Lumina具身智能社区] 具身智能技术指南 Embodied-AI-Guide

spoc-robot-training. SPOC: Imitating Shortest Paths in Simulation Enables Effective Navigation and Manipulation in the Real World

vggt. [CVPR 2025 Best Paper Award] VGGT: Visual Geometry Grounded Transformer

PredDRL. Python

gym-hil. Human in the loop Reinforcement Learning suite

VLA-Diffusion-Policy-Robotics. Awesome collection of resources and papers on Diffusion Models for Robotic Manipulation.

Awesome-VLA-RL. This repository summarizes recent advances in the VLA + RL paradigm and provides a taxonomic classification of relevant works.

PRIME. Scalable RL solution for advanced reasoning of language models

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

ReinFlow. [NeurIPS 2025] Flow x RL. "ReinFlow: Fine-tuning Flow Policy with Online Reinforcement Learning". Support VLAs e.g., Pi0, Pi0.5, GR00TN1.5. Fully open-sourced.

hume. 🦾 A Dual-System VLA with System2 Thinking

meshcat. Remotely-controllable 3D viewer, built on top of three.js

RL4VLA. Python

boring.notch. TheBoringNotch: Not so boring notch That Rocks 🎸🎶

hil-serl-sim. This warehouse has been improved based on hil-serl, adding simulation training functionality

verl. verl/HybridFlow: A Flexible and Efficient RL Post-Training Framework

SimpleVLA-RL. [ICLR 2026] SimpleVLA-RL: Scaling VLA Training via Reinforcement Learning

AutoBio. Preliminary version of AutoBio (https://arxiv.org/abs/2505.14030)

RoboMani-PaperList. A Paper List for Robotic Manipulation.

conrft. This is the official implementation of the paper "ConRFT: A Reinforced Fine-tuning Method for VLA Models via Consistency Policy".

PARL. A high-performance distributed training framework for Reinforcement Learning

robotwin. Python

Chemistry3D. Python

umi-on-legs. UMI on Legs: Making Manipulation Policies Mobile with Manipulation-Centric Whole-body Controllers

RAGEN. RAGEN leverages reinforcement learning to train LLM reasoning agents in interactive, stochastic environments.

RoboVerse. RoboVerse: Towards a Unified Platform, Dataset and Benchmark for Scalable and Generalizable Robot Learning

cleanrl. High-quality single file implementation of Deep Reinforcement Learning algorithms with research-friendly features (PPO, DQN, C51, DDPG, TD3, SAC, PPG)

humanoid-bench. Python

CLIP. CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image

GR-1. Code for "Unleashing Large-Scale Video Generative Pre-training for Visual Robot Manipulation"

auto_eval. AutoEval: Autonomous Evaluation of Generalist Robot Manipulation Policies in the Real World | CoRL 2025

TesserAct. ICCV 2025 | TesserAct: Learning 4D Embodied World Models

collector. [ICRA 2024] AirExo: Low-Cost Exoskeletons for Learning Whole-Arm Manipulation in the Wild

LAPA. [ICLR 2025] LAPA: Latent Action Pretraining from Videos

Metaworld. Collections of robotics environments geared towards benchmarking multi-task and meta reinforcement learning

pyroki. A Modular Toolkit for Robot Kinematic Optimization

XLeRobot. XLeRobot: Practical Dual-Arm Mobile Home Robot for $660

vlarl. Single-file implementation to advance vision-language-action (VLA) models with reinforcement learning.

Improved-3D-Diffusion-Policy. [IROS 2025] Generalizable Humanoid Manipulation with 3D Diffusion Policies. Part 1: Train & Deploy of iDP3

DemoGen. [RSS25] Official implementation of DemoGen: Synthetic Demonstration Generation for Data-Efficient Visuomotor Policy Learning

discover-hidden-visual-concepts. Official implementation of `Discovering Hidden Visual Concepts Beyond Linguistic Input in Infant Learning`, CVPR 2025

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

mshab. A Benchmark for Low-Level Manipulation in Home Rearrangement Tasks

vres. Vacation Research Experience Topic: Developing Bimanual Teleoperation Device for Dual Robotic Manipulator Arms and Robot Learning

triad_openvr. This is an enhanced wrapper for the already excellent pyopenvr library by cmbruns. The goal of this library is to create easy to use python functions for any SteamVR tracked system.

r3m. Pre-training Reusable Representations for Robotic Manipulation Using Diverse Human Video Data

vr-teleoperation. Code for bimanual teleoperation with Meta Quest3

3d-foundation-policy. Jupyter Notebook

PointFlowMatch. Jupyter Notebook

EAGLE. Official Implementation of EAGLE-1 (ICML'24), EAGLE-2 (EMNLP'24), and EAGLE-3 (NeurIPS'25).

TSP3D. [CVPR 2025, All Strong Accept] TSP3D: Text-guided Sparse Voxel Pruning for Efficient 3D Visual Grounding

calvin. CALVIN - A benchmark for Language-Conditioned Policy Learning for Long-Horizon Robot Manipulation Tasks

brs-ctrl. Official Hardware Codebase for the Paper "BEHAVIOR Robot Suite: Streamlining Real-World Whole-Body Manipulation for Everyday Household Activities"

brs-algo. Official Algorithm Codebase for the Paper "BEHAVIOR Robot Suite: Streamlining Real-World Whole-Body Manipulation for Everyday Household Activities"

furniture-bench. FurnitureBench: Real-World Furniture Assembly Benchmark (RSS 2023)

Slow_Thinking_with_LLMs. A series of technical report on Slow Thinking with LLM

STP. stp files of Tabletop Teleoperation structures

Product_User_Guide. Config files for my GitHub profile.

anyplace. Official implementation of "AnyPlace: Learning Generalized Object Placement for Robot Manipulation"

RLBench. A large-scale benchmark and learning environment.

LIBERO. Benchmarking Knowledge Transfer in Lifelong Robot Learning

dift. [NeurIPS'23] Emergent Correspondence from Image Diffusion

V-GPS. official implementation for our paper Steering Your Generalists: Improving Robotic Foundation Models via Value Guidance (CoRL 2024)

nerfies.github.io. JavaScript

Awesome-Robotics-Diffusion. A curated list of recent robot learning papers incorporating diffusion models for robotics tasks.