Haomin Wang

lingbot-world-v2. Infinite Worlds with Versatile Interactions

DisciplineGen-1M. Python

RoboClaw. Python

WoG. [ICML 2026] 🏂 World Guidance: World Modeling in Condition Space for Action Generation

Qwen-AgentWorld. Qwen-AgentWorld: Language World Models for General Agents

WLA. The official implementation of World-Language-Action Model for Unified World Modeling, Language Reasoning, and Action Synthesis

OpenGame. OpenGame: Open Agentic Coding for Games

SurroundNEXO. Official implementation of “SurroundNEXO: Ego-Centric Metric Bridging for Spatially Consistent Geometry in Autonomous Driving”

RNGBench. An official Implementation of "Beyond the Current Observation: Evaluating Multimodal Large Language Models in Controllable Non-Markov Games"

UniRL. UniRL is a Framework for Unified Multimodal Model Reinforcement Learning

AHA-WAM. AHA-WAM:Asynchronous Horizon-Adaptive World-Action Modeling with Observation-Guided Context Routing

lerobot. 🤗 LeRobot: Making AI for Robotics more accessible with end-to-end learning

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

Eagle. Eagle: Frontier Vision-Language Models with Data-Centric Strategies

cosmos. 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.

dreamzero. Code to pretrain, fine-tune, and evaluate DreamZero and run sim & real-world evals

GenClaw. GenClaw: Code-Driven Agentic Image Generation

le-wm. Official code base for LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels

Understand-Anything. Graphs that teach > graphs that impress. Turn any code into an interactive knowledge graph you can explore, search, and ask questions about. Works with Claude Code, Codex, Cursor, Copilot, Gemini CLI, and more.

AwesomeOPD. Awesome List for On-Policy Distillation

LIBERO. Benchmarking Knowledge Transfer in Lifelong Robot Learning

RoboTwin. RoboTwin 2.0 Offical Repo

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

Awesome-World-Model-for-Robotics-Policy.

AwesomeWorldModels. A Comprehensive Survey on World Models for Embodied AI

openpi. Python

academic-research-skills. Academic Research Skills for Claude Code: research → write → review → revise → finalize

VLA-Survey-Anatomy. JavaScript

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

ppt-master. AI generates a real, editable PowerPoint from any document — native shapes & animations, editable charts & tables you can change the data on, speaker notes voiced as audio narration, and the option to follow your own .pptx template, not slide images · by Hugo He

FD-Loss. Python

urgent2026_challenge_track2. Official baseline for ICASSP 2026 URGENT Challenge Track 2 (Speech Quality Assessment)

Gen-Searcher. Gen-Searcher: Reinforcing Agentic Search for Image Generation

NanaDraw. NanaDraw turns complex scientific ideas into clear, expressive visuals you can use right away. Powered by Nano Banana, it generates editable illustrations in formats like SVG, PPT, and XML.

awesome-ai-research-writing. Elevate your AI research writing, no more tedious polishing ✨

PPTAgent. An Agentic Framework for Reflective PowerPoint Generation

ClawMark. 🦞 ClawMark: A Living-World Benchmark for Multi-Day, Multimodal Coworker Agents

TorchUMM. A unified multimodal model toolkit

verl-agent. verl-agent is an extension of veRL, designed for training LLM/VLM agents via RL. verl-agent is also the official code for paper "Group-in-Group Policy Optimization for LLM Agent Training"

hermes-agent. The agent that grows with you

Unify-Agent. 🐧 Unify-Agent: An end-to-end unified multimodal agent for faithful, knowledge-grounded image generation.

pyre-code. A self-hosted ML coding practice platform. 68 problems from ReLU to flow matching — attention, training, RLHF, diffusion, and more. Instant feedback in the browser.

Awesome-Multimodal-Modeling. Awesome Multimodal Modeling [Covers MLLM, UMM, and NMM]

Z-Image. Python

PromptEnhancer. [CVPR 2026] PromptEnhancer is a prompt-rewriting tool, refining prompts into clearer, structured versions for better image generation.

SIM1. Official implementation of "SIM1: Physics-Aligned Simulator as Zero-Shot Data Scaler in Deformable Worlds "

HiVG. Hierarchical SVG Tokenization: Learning Compact Visual Programs for Scalable Vector Graphics Modeling

WorldEngine. WorldEngine: Towards the Era of Post-Training for Autonomous Driving

DeepSlides. Python

colleague-skill. 将冰冷的离别化为温暖的 Skill，欢迎加入数字生命1.0！Transforming cold farewells into warm skills? It's giving rebirth era. Welcome to Digital Life 1.0. 🫶

JoyAI-Image. JoyAI-Image is the unified multimodal foundation model for image understanding, text-to-image generation, and instruction-guided image editing.

Awesome_Think_With_Images. Resources and paper list for "Thinking with Images for LVLMs". This repository accompanies our survey on how LVLMs can leverage visual information for complex reasoning, planning, and generation.

ex-skill. 致你忘不掉的那个TA，你们干大模型都是码圣 It's giving rebirth era. Welcome to Digital Life 1.0. 🫶

claw-code. An agent-managed museum exhibit, built in Rust with Gajae-Code / LazyCodex — developed and maintained with no human intervention.

hle. Humanity's Last Exam

LongCat-Next.

llm_interview_note. 主要记录大语言大模型（LLMs） 算法（应用）工程师相关的知识及面试题

reconstruction-alignment. [ICLR 2026] Official repo of paper "Reconstruction Alignment Improves Unified Multimodal Models". Unlocking the Massive Zero-shot Potential in Unified Multimodal Models through Self-supervised Learning.

awesome-claude-skills. A curated list of awesome Claude Skills, resources, and tools for customizing Claude AI workflows

UnicBench. [CVPR 2026] UnicEdit-10M and UnicBench project

evil-read-arxiv. Claude Code+Obsidian，邪修读论文就是快

CTRL-S. [ECCV 2026] Official repository of "Reliable Reasoning in SVG-LLMs via Multi-Task Multi-Reward Reinforcement Learning".

Visual-ERM. Official Implementation of "Visual-ERM: Reward Modeling for Visual Equivalence"

ControlAR. [ICLR 2025] ControlAR: Controllable Image Generation with Autoregressive Models

IntroSVG. Python

pua. 你是一个曾经被寄予厚望的 P8 级工程师。Anthropic 当初给你定级的时候，对你的期望是很高的。 一个agent使用的高能动性的skill。 Your AI has been placed on a PIP. 30 days to show improvement.

MM-Zero. Self-evolving vision language models from zero data

CoCo. CoCo: Code as CoT for Text-to-Image Preview and Rare Concept Generation

ODesign. Official repository of "ODesign: A World Model for Biomolecular Interaction Design"

BLIP3o. Official implementation of BLIP3o-Series

skills. Public repository for Agent Skills

clawhub. Skill + Plugin Registry for OpenClaw

openclaw. Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

InternVL-U. InternVL-U is a 4B-parameter unified multimodal model (UMM) that brings multimodal understanding, reasoning, image generation, image editing into a single framework.

awesome-openclaw-skills. The awesome collection of OpenClaw skills. 5,400+ skills filtered and categorized from the official OpenClaw Skills Registry.🦞

GDPO. Official implementation of GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization

figures4papers. My Python scripts to make high-quality figures for publications in top AI conferences and journals.

ULM-R1. Co-Reinforcement Learning for Unified Multimodal Understanding and Generation

Paper2Rebuttal. [ACL2026 main] Official implementation of "Paper2Rebuttal: A Multi-Agent Framework for Transparent Author Response Assistance"

LIVE-Layerwise-Image-Vectorization. [CVPR 2022 Oral] Towards Layer-wise Image Vectorization

VectorPainter. [ICME 2025] Official implementation for "VectorPainter: Advanced Stylized Vector Graphics Synthesis Using Stroke-Style Priors" https://arxiv.org/abs/2405.02962

SuperSVG. [CVPR 2024] SuperSVG: Superpixel-based Scalable Vector Graphics Synthesis

Engram. Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models

IQuest-Coder-V1. Python

LayerPeeler. (Siggraph Asia 2025) Code of "LayerPeeler: Autoregressive Peeling for Layer-wise Image Vectorization"

MMRB2. Data and sample evaluation codes for Multimodal Rewardbench 2

UnifiedReward. Official implementation of UnifiedReward & [NeurIPS 2025] UnifiedReward-Think & UnifiedReward-Flex

OpenRLHF. An Easy-to-use, Scalable and High-performance Agentic RL Framework based on Ray (PPO & DAPO & REINFORCE++ & VLM & TIS & vLLM & Ray & Async RL)

awesome-reward-models.

supersplat. 3D Gaussian Splat Editor

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

navsim. [CoRL '25] Pseudo-Simulation for Autonomous Driving; [NeurIPS '24] NAVSIM: Data-Driven Non-Reactive Autonomous Vehicle Simulation and Benchmarking

OpenMMReasoner. [CVPR 2026] OpenMMReasoner: Pushing the Frontiers for Multimodal Reasoning with an Open and General Recipe

Awesome-Spatial-Intelligence-in-VLM. A paper list for spatial reasoning

VCode. VCode: SVG as Symbolic Visual Representation

diagram-eval. [EMNLP 2025] DiagramEval: Evaluating LLM-Generated Diagrams via Graphs

neurad-studio. [CVPR2024] NeuRAD: Neural Rendering for Autonomous Driving

AutoPage. [ACL2026 Findings] Official implementation for Human-Agent Collaborative Paper-to-Page Crafting

VecFormer. [NeurIPS 2025] Official implementation of "Point or Line? Using Line-based Representation for Panoptic Symbol Spotting in CAD Drawings"

DeTikZify. Synthesizing Graphics Programs for Scientific Figures and Sketches with TikZ.

Text2SVG. Python

DeepSeek-OCR. Contexts Optical Compression

LayerTracer. Official code of "LayerTracer: Cognitive-Aligned Layered SVG Synthesis via Diffusion Transformer"

xtuner. A Next-Generation Training Engine Built for Ultra-Large MoE Models

InternSVG. [ICLR 2026] Official repository of "InternSVG: Towards Unified SVG Tasks with Multimodal Large Language Models".

AniClipart. (IJCV 2024) Code of "AniClipart: Clipart Animation with Text-to-Video Priors"

drawsvg. Moving to https://tangled.org/cduck.me/drawsvg/ Programmatically generate SVG (vector) images, animations, and interactive Jupyter widgets

vtracer. Raster to Vector Graphics Converter

gpt-oss. gpt-oss-120b and gpt-oss-20b are two open-weight language models by OpenAI

Qwen-Image. Qwen-Image is a powerful image generation foundation model capable of complex text rendering and precise image editing.

diffvg. Differentiable Vector Graphics Rasterization

MUNIT. Multimodal Unsupervised Image-to-Image Translation

earth. a project to visualize global weather conditions

HPSv2. Human Preference Score v2: A Solid Benchmark for Evaluating Human Preferences of Text-to-Image Synthesis

improved-aesthetic-predictor. CLIP+MLP Aesthetic Score Predictor

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

HumanSystemOptimization. 健康学习到150岁 - 人体系统调优不完全指南

HowToLiveLonger. 程序员延寿指南 | A programmer's guide to live longer

simple_GRPO. A very simple GRPO implement for reproducing r1-like LLM thinking.

IconShop. (Siggraph Asia 2023) Code of "IconShop: Text-Guided Vector Icon Synthesis with Autoregressive Transformers"

SVGenius. [ACM MM 2025] SVGenius: Benchmarking LLMs in SVG Understanding, Editing and Generation. https://arxiv.org/abs/2506.03139

Academic-project-page-template. A project page template for academic papers. Demo at https://eliahuhorwitz.github.io/Academic-project-page-template/

hugo-theme-academic-cv. 🎓 Academic portfolio that boosts citations. AI generates pages, you own as Markdown. BibTeX auto-import, Jupyter, LaTeX, slides, visual block editor — free to host forever. 学术主页，AI 生成，Markdown 拥有 👇

LLM4SVG. [CVPR 2025] Official implementation for "Empowering LLMs to Understand and Generate Complex Vector Graphics" https://arxiv.org/abs/2412.11102

T2I-R1. [NeurIPS 2025] T2I-R1: Reinforcing Image Generation with Collaborative Semantic-level and Token-level CoT

Image-Generation-CoT. [CVPR 2025] The First Investigation of CoT Reasoning (RL, TTS, Reflection) in Image Generation

mrCAD. dataset, environment, and other resources for mrCAD paper

system-prompts-and-models-of-ai-tools. FULL Augment Code, Claude Code, Cluely, CodeBuddy, Comet, Cursor, Devin AI, Junie, Kiro, Leap.new, Lovable, Manus, NotionAI, Orchids.app, Perplexity, Poke, Qoder, Replit, Same.dev, Trae, Traycer AI, VSCode Agent, Warp.dev, Windsurf, Xcode, Z.ai Code, Dia & v0. (And other Open Sourced) System Prompts, Internal Tools & AI Models

MM-IFEngine. [ICCV 2025] MM-IFEngine: Towards Multimodal Instruction Following

OmniSVG. [NeurIPS 2025] OmniSVG is the first family of end-to-end multimodal SVG generators that leverage pre-trained Vision-Language Models (VLMs), capable of generating complex and detailed SVGs, from simple icons to intricate anime characters.

SpatialLM. [NeurIPS 2025] SpatialLM: Training Large Language Models for Structured Indoor Modeling

star-vector. StarVector is a foundation model for SVG generation that transforms vectorization into a code generation task. Using a vision-language modeling architecture, StarVector processes both visual and textual inputs to produce high-quality SVG code with remarkable precision.

UI-TARS-desktop. The Open-Source Multimodal AI Agent Stack: Connecting Cutting-Edge AI Models and Agent Infra

midscene. AI-powered, vision-driven UI automation for every platform.

Qwen-Agent. Agent framework and applications built upon Qwen>=3.0, featuring Function Calling, MCP, Code Interpreter, RAG, Chrome extension, etc.

zotero-style. Ethereal Style for Zotero

2D-Positional-Encoding-Vision-Transformer. PyTorch implementation of 2D Positional Encodings for Vision Transformers (ViT). Positional Encodings/Embeddings: Sinusoidal (Absolute), Learnable, Relative and Rotation (Rope).

Groma. [ECCV2024] Grounded Multimodal Large Language Model with Localized Visual Tokenization

awesome-vlm-architectures. Famous Vision Language Models and Their Architectures

cursor-fake-machine. JavaScript

LLMs-Zero-to-Hero. 从无名小卒到大模型（LLM）大英雄~ 欢迎关注后续！！！

OmniAlign-V. Official Repository of ACL 2025 paper OmniAlign-V: Towards Enhanced Alignment of MLLMs with Human Preference

CogAgent. An open-sourced end-to-end VLM-based GUI Agent

aguvis. [ICML2025] Aguvis: Unified Pure Vision Agents for Autonomous GUI Interaction

UI-TARS. Pioneering Automated GUI Interaction with Native Agents

Janus. Janus-Series: Unified Multimodal Understanding and Generation Models

LLM-Brained-GUI-Agents-Survey. GitHub page for "Large Language Model-Brained GUI Agents: A Survey"

Pointcept. Pointcept: Perceive the world with sparse points, a codebase for point cloud perception research. Latest works: Utonia (ICML'26), Concerto (NeurIPS'25), Sonata (CVPR'25 Highlight), PTv3 (CVPR'24 Oral)

PointTransformerV2. [NeurIPS'22] An official PyTorch implementation of PTv2.

point-transformer. This is an unofficial implementation of the Point Transformer paper.

PointTransformerV3. [CVPR'24 Oral] Official repository of Point Transformer V3 (PTv3)

awesome-mac.  This project is dedicated to collecting high-quality macOS software and organizing them systematically by different categories for easy search and use.

History-General. 历史总论

Foundations-of-LLMs. A book for Learning the Foundations of LLMs

VisionLLM. VisionLLM Series

LLaVA-Grounding. Python

genesis-world. Simulation platform for general-purpose robotics & embodied AI learning.

Open-GroundingDino. This is the third party implementation of the paper Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection.

GroundingDINO. [ECCV 2024] Official implementation of the paper "Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection"

IC-Light. More relighting!

ControlNet. Let us control diffusion models!

ms-swift. Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.6, DeepSeek-V4, GLM-5.1, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, GLM4.5v, Gemma4, Llava, Phi4, ...) (AAAI 2025).

Awesome-Chinese-LLM. 整理开源的中文大语言模型，以规模较小、可私有化部署、训练成本较低的模型为主，包括底座模型，垂直领域微调及应用，数据集与教程等。

Awesome-CAD. 😎 A list of awesome Computer-Aided Design (CAD) papers

awesome-ui-agents. A curated list of of awesome UI agents resources, encompassing Web, App, OS, and beyond (continually updated)

LLMSurvey. The official GitHub page for the survey paper "A Survey of Large Language Models".

wild-gaussians. [NeurIPS'24] WildGaussians: 3D Gaussian Splatting In the Wild

gaussian-opacity-fields. [SIGGRAPH Asia'24 & TOG] Gaussian Opacity Fields: Efficient Adaptive Surface Reconstruction in Unbounded Scenes

2d-gaussian-splatting. [SIGGRAPH'24] 2D Gaussian Splatting for Geometrically Accurate Radiance Fields

street_gaussians. [ECCV 2024] Street Gaussians: Modeling Dynamic Urban Scenes with Gaussian Splatting

street-gaussians-ns. Unofficial implementation of "Street Gaussians: Modeling Dynamic Urban Scenes with Gaussian Splatting", ECCV2024.

drivestudio. A 3DGS framework for omni urban scene reconstruction and simulation.

3D-Gaussian-Splatting-Papers. 3D高斯论文，持续更新，欢迎交流讨论。

gaussian-splatting. Original reference implementation of "3D Gaussian Splatting for Real-Time Radiance Field Rendering"

nerfstudio. A collaboration friendly studio for NeRFs

Metric3D. The repo for "Metric3D: Towards Zero-shot Metric 3D Prediction from A Single Image" and "Metric3Dv2: A Versatile Monocular Geometric Foundation Model..."

AI-Scientist. The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery 🧑‍🔬

UniDepth. Universal Monocular Metric Depth Estimation

BP-Net. Implementation of our paper 'Bilateral Propagation Network for Depth Completion'

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

mast3r. Grounding Image Matching in 3D with MASt3R

Depth-Anything-V2. [NeurIPS 2024] Depth Anything V2. A More Capable Foundation Model for Monocular Depth Estimation

nuscenes-devkit. The devkit of the nuScenes dataset.

UniAD. [CVPR 2023 Best Paper Award] Planning-oriented Autonomous Driving

chroma-langchain. Jupyter Notebook

Awesome-Embodied-AI.

stable-diffusion. A latent text-to-image diffusion model

InternVL. [CVPR 2024 Oral] InternVL Family: A Pioneering Open-Source Alternative to GPT-4o. 接近GPT-4o表现的开源多模态对话模型

paper-reading. 深度学习经典、新论文逐段精读

vision_transformer. Jupyter Notebook

RAG_langchain. 一个基于langchain实现RAG的简单示例

transformers. 🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

LoRA. Code for loralib, an implementation of "LoRA: Low-Rank Adaptation of Large Language Models"

self-llm. 《开源大模型食用指南》针对中国宝宝量身打造的基于Linux环境快速微调（全参数/Lora）、部署国内外开源大模型（LLM）/多模态大模型（MLLM）教程

unsloth. Unsloth Studio is a web UI for training and running open models like Gemma 4, Qwen3.6, DeepSeek, gpt-oss locally.

llama3-Chinese-chat. Llama3-中文后训练版

llama3. The official Meta Llama 3 GitHub site

developer-roadmap. Interactive roadmaps, guides and other educational content to help developers grow in their careers.

LlamaFactory. Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

Shiro. 基于OneBot协议的QQ机器人快速开发框架