Yuchen Lin

PMT. Official implementation of Perceptive Behavior Foundation Model

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.

pymupdf4llm. PyMuPDF4LLM

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

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

browser-agent. [browser-agent] Never send a human to do a machine's job.

pku-undergraduate-thesis. TeX

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

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

fpo-control. Flow Policy Gradients for Robot Control

Mahjong-B2D. 日麻从入门到入土，相关新手入门教程和指南书籍

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

Causal-Forcing. [ICML 2026] Official codebase for "Causal Forcing: Autoregressive Diffusion Distillation Done Right for High-Quality Real-Time Interactive Video Generation" & Causal Forcing++

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

lingbot-world. Advancing Open-source World Models

cosmos-policy. Cosmos Policy

LIBERO. Benchmarking Knowledge Transfer in Lifelong Robot Learning

Action100M. A Large-scale Video Action Dataset

openpi. Python

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

openevolve. Open-source implementation of AlphaEvolve

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

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.

NitroGen. A Foundation Model for Generalist Gaming Agents

GPT-Rewind. Python

TRELLIS.2. Native and Compact Structured Latents for 3D Generation

BindWeave. [ICLR 2026] Official Repo For "BindWeave: Subject-Consistent Video Generation via Cross-Modal Integration"

JarvisEvo. [CVPR' 2026] JarvisEvo: Towards a Self-Evolving Photo Editing Agent with Synergistic Editor-Evaluator Optimization

particulate. [CVPR 2026] Official Implementation of Particulate: Feed-Forward 3D Object Articulation

Diff4Splat. [CVPR 2026] Diff4Splat: Controllable 4D Scene Generation with Latent Dynamic Reconstruction Models.

openvla. OpenVLA: An open-source vision-language-action model for robotic manipulation.

VITRA. [ICRA 2026] VITRA: Scalable Vision-Language-Action Model Pretraining for Robotic Manipulation with Real-Life Human Activity Videos

DiffSynth-Studio. Enjoy the magic of Diffusion models!

CoVT. [ECCV 2026] Official repo of "Chain-of-Visual-Thought: Teaching VLMs to See and Think Better with Continuous Visual Tokens"

ARPO. [ICLR 2026] Agentic Reinforced Policy Optimization (ARPO)

FFGO-Video-Customization. Video Content Customization Using First Frame

ToolRL. Python

CS-PhD-Application-fee-waivers. Collections of CS PhD Application Fee Waivers of schools in North America

MambaVision. [CVPR 2025] Official PyTorch Implementation of MambaVision: A Hybrid Mamba-Transformer Vision Backbone

Lecture-Note-PKU-Machine-Learning-25Fall. A lecture note of Prof. Liwei Wang's course on Machine Learning.

CAST. An unofficial and simplified implementation of SIGGRAPH 2025 best paper nominate: CAST: Component-Aligned 3D Scene Reconstruction from an RGB Image, working in progress

DanceGRPO. An official implementation of DanceGRPO: Unleashing GRPO on Visual Generation

BranchGRPO. BranchGRPO: Stable and Efficient GRPO with Structured Branching in Diffusion Models

VideoREPA. [NeurIPS 2025] VideoREPA: Learning Physics for Video Generation through Relational Alignment with Foundation Models

Sparse-VideoGen. [ICML2025, NeurIPS2025 Spotlight] Sparse VideoGen 1 & 2: Accelerating Video Diffusion Transformers with Sparse Attention

HoloCine. [CVPR 2026 Highlight] Official Implementations for Paper - HoloCine: Holistic Generation of Cinematic Multi-Shot Long Video Narratives

Qwen3-VL. Qwen3-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.

REPA. [ICLR'25 Oral] Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think

dinov3. Reference PyTorch implementation and models for DINOv3

VideoNSA. VideoNSA: Native Sparse Attention Scales Video Understanding

RAE. Official PyTorch Implementation of "Diffusion Transformers with Representation Autoencoders"

Self-Forcing. Official codebase for "Self Forcing: Bridging Training and Inference in Autoregressive Video Diffusion" (NeurIPS 2025 Spotlight)

Infinite-Forcing. Infinite-Forcing: Towards Infinite-Long Video Generation

LongLive. Long Video Gen Infrastructure

tinyworlds. A minimal implementation of DeepMind's Genie world model

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

OmniGen2. OmniGen2: Exploration to Advanced Multimodal Generation. https://arxiv.org/abs/2506.18871

Hunyuan-GameCraft-1.0. Hunyuan-GameCraft: High-dynamic Interactive Game Video Generation with Hybrid History Condition

RLFromScratch. Python

MeshArt. MeshArt: Generating Articulated Meshes with Structure-Guided Transformers (CVPR2025)

Wan2.2. Wan: Open and Advanced Large-Scale Video Generative Models

InteriorGS. InteriorGS: 3D Gaussian Splatting Dataset of Semantically Labeled Indoor Scenes

HunyuanWorld-1.0. Generating Immersive, Explorable, and Interactive 3D Worlds from Words or Pixels with Hunyuan3D World Model

co3d. Tooling for the Common Objects In 3D dataset.

ObjectGS. [ICCV 2025] ObjectGS: Object-aware Scene Reconstruction and Scene Understanding via Gaussian Splatting

Pi3. [ICLR 2026] π^3: Permutation-Equivariant Visual Geometry Learning

getting-started-handout. 北京大学计算机基础能力手册

PartCrafter-Windows. PartCrafter: Structured 3D Mesh Generation via Compositional Latent Diffusion Transformers for Windows.

AutoPartGen.

SpaTrackerV2. [ICCV 2025] SpatialTrackerV2: 3D Point Tracking Made Easy

PhysX-3D. PhysX: Physical-Grounded 3D Asset Generation (NeurIPS 2025, Spotlight)

MoVieS. [CVPR 2026] Official implementation of "MoVieS: Motion-Aware 4D Dynamic View Synthesis in One Second".

copart. CoPart (ICCV 2025): A part-based 3D generation framework & the first large-scale part-level 3D dataset.

PartCrafter. [NeurIPS 2025] PartCrafter: Structured 3D Mesh Generation via Compositional Latent Diffusion Transformers

tilelang. Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels

ccf-deadlines. ⏰ Agenticly track worldwide conference deadlines (Website, Python Cli, Wechat Applet)

OpenThinkIMG. OpenThinkIMG is an end-to-end open-source framework that empowers LVLMs to think with images.

SimpleAR. Pytorch implementation for the paper titled "SimpleAR: Pushing the Frontier of Autoregressive Visual Generation"

MediaCrawler. 小红书笔记 | 评论爬虫、抖音视频 | 评论爬虫、快手视频 | 评论爬虫、B 站视频 ｜ 评论爬虫、微博帖子 ｜ 评论爬虫、百度贴吧帖子 ｜ 百度贴吧评论回复爬虫 | 知乎问答文章｜评论爬虫

alltracker. AllTracker is a model for tracking all pixels in a video.

Ovis-U1. An unified model that seamlessly integrates multimodal understanding, text-to-image generation, and image editing within a single powerful framework.

Bagel. Open-source unified multimodal model

TileOPs. High-performance LLM operator library built on TileLang.

test-time-registers. [NeurIPS '25 Spotlight] Official Pytorch implementation of "Vision Transformers Don't Need Trained Registers"

assembler3d. Code release of "Assembler: Scalable 3D Part Assembly via Anchor Point Diffusion".

AIProgramming-Project. My project for the course 'AI Programming' in Jan. 2025.

metaquery. Official Implementation of Paper Transfer between Modalities with MetaQueries

JarvisArt. [NeurIPS' 2025] JarvisArt: Liberating Human Artistic Creativity via an Intelligent Photo Retouching Agent

octotools. OctoTools: An agentic framework with extensible tools for complex reasoning

HARL. Official implementation of HARL algorithms based on PyTorch.

Introduction_to_Database_Systems-clj-2024Spring-PKU. I love Database.

Hunyuan3D-2.1. From Images to High-Fidelity 3D Assets with Production-Ready PBR Material

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

Sparc3D. Official repo for paper "Sparse Representation and Construction for High-Resolution 3D Shapes Modeling".

cosmos-transfer1-diffusion-renderer. Cosmos-Transfer1-DiffusionRenderer: High-quality video de-lighting and re-lighting based on Cosmos video diffusion framework

kyvo. This repository contains the code for the paper - "Aligning Text, Images, and 3D Structure Token-by-Token" (CVPR 2026)

PartPacker. Efficient Part-level 3D Object Generation via Dual Volume Packing

ShapeLLM-Omni. [NeurIPS 2025 Spotlight] A Native Multimodal LLM for 3D Generation and Understanding

BeautifulPointCloud. Transform your point cloud data into beautifully rendered 3D images.

latex_paper_writing_tips. Tips for Writing a Research Paper using LaTeX

MV-Adapter. [ICCV 2025] Official impl. of "MV-Adapter: Multi-view Consistent Image Generation Made Easy"

BLIP3o. Official implementation of BLIP3o-Series

SAMPart3D. SAMPart3D: Segment Any Part in 3D Objects

rectified-flow-pytorch. Implementation of rectified flow and some of its followup research / improvements in Pytorch

Hunyuan3D-2. High-Resolution 3D Assets Generation with Large Scale Hunyuan3D Diffusion Models.

rewarped. Reinforcement learning in differentiable multiphysics simulation with NVIDIA Warp.

ObjaversePlusPlus. [ICCV 2025] Repo for Objaverse++, Curated 3D Object Dataset with Quality Annotations

HoloPart. HoloPart: Generative 3D Part Amodal Segmentation

part123. https://liuar0512.github.io/part123_official_page/

TRELLIS. Official repo for paper "Structured 3D Latents for Scalable and Versatile 3D Generation" (CVPR'25 Spotlight).

TripoSG. TripoSG: High-Fidelity 3D Shape Synthesis using Large-Scale Rectified Flow Models

trl. Train transformer language models with reinforcement learning.

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

dso. [ICCV 2025] DSO: Aligning 3D Generators with Simulation Feedback for Physical Soundness

EasyR1. EasyR1: An Efficient, Scalable, Multi-Modality RL Training Framework based on veRL

ShapeSplat-Gaussian_MAE. [3DV 2025 Oral]: A Large-scale Dataset of Gaussian Splats and Their Self-Supervised Pretraining

servers. Model Context Protocol Servers

PartRM. [CVPR 2025] Offical Code Implementation for PartRM

supersplat. 3D Gaussian Splat Editor

VLM-R1. Solve Visual Understanding with Reinforced VLMs

MetaSpatial. [ICLR 2026] MetaSpatial leverages reinforcement learning to enhance 3D spatial reasoning in vision-language models (VLMs), enabling more structured, realistic, and adaptive scene generation for applications in the metaverse, AR/VR, and game development.

stanford-shapenet-renderer. Scripts for batch rendering models using Blender. Tested with models from stanfords shapenet library.

Awesome-Articulated-Object-Understanding. A curated list of resources for articulated objects understanding.

NeuMA. [NeurIPS 2024] NeuMA: Neural Material Adaptor for Visual Grounding of Intrinsic Dynamics

Awesome-Optical-Flow. This is a list of awesome paper about optical flow and related work.

SEA-RAFT. [ECCV2024 - Oral, Best Paper Award Candidate] SEA-RAFT: Simple, Efficient, Accurate RAFT for Optical Flow

MPM-PyTorch. A PyTorch Implementation of MLS-MPM (Moving Least Squares Material Point Method)

torch-batch-svd. A 100x faster SVD for PyTorch⚡️

awesome-deep-reasoning. Collect every awesome work about r1!

TreeMeshGPT. [CVPR 2025] TreeMeshGPT: Artistic Mesh Generation with Autoregressive Tree Sequencing

awesome-simulation. Resources for Physics based simulation in Computer Graphics 图形学中物理模拟的资源整理

taichi_mpm. High-performance moving least squares material point method (MLS-MPM) solver. (ACM Transactions on Graphics, SIGGRAPH 2018)

blender-mcp. Open-source MCP to use Blender with any LLM

Awesome-Physically-Interactive-World-Models. A curated list of papers and open-source resources focused on Physics-Inspired 3D Reconstruction and Simulation, intended to keep pace with the anticipated surge of research in the coming months. If you have any additions or suggestions, feel free to contribute. Additional resources like blog posts, videos, etc. are also welcome.

common_metrics_on_video_quality. You can easily calculate FVD, PSNR, SSIM, LPIPS for evaluating the quality of generated or predicted videos.

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

llama. Inference code for Llama models

LLaMA-Mesh. Unifying 3D Mesh Generation with Language Models

urdformer. code release for URDFormer

singapo. [ICLR 2025] SINGAPO: Single Image Controlled Generation of Articulated Parts in Objects

real2code. Python

MagicArticulate. [CVPR 2025] Official repository for “MagicArticulate: Make Your 3D Models Articulation-Ready”

gic. Gaussian-Informed Continuum for Physical Property Identification and Simulation

PAC-NeRF. Physics Augmented Continuum Neural Radiance Fields for Geometry-Agnostic System Identification

latent-radiance-field. [ICLR 2025] Latent Radiance Fields with 3D-aware 2D Representations

articulate-anything. [ICLR 2025] Official implementation of Articulate-Anything

Convex-Analysis-homework. 2021 Spring

RoboGen. A generative and self-guided robotic agent that endlessly propose and master new skills.

DeepSeek-R1.

PKUElective2022Spring. PKUAutoElective for 2022 spring semester

omniphysgs. [ICLR 2025] OmniPhysGS: 3D Constitutive Gaussians for General Physics-based Dynamics Generation

DiffSplat. [ICLR 2025] Official implementation of "DiffSplat: Repurposing Image Diffusion Models for Scalable 3D Gaussian Splat Generation".

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

DepthCrafter. [CVPR 2025 Highlight] DepthCrafter: Generating Consistent Long Depth Sequences for Open-world Videos

physics-IQ-benchmark. Benchmarking physical understanding in generative video models

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

humansplat. [NeurIPS'24] Official implementation of "HumanSplat: Generalizable Single-Image Human Gaussian Splatting with Structure Priors"

dmesh. Official implementation for "DMesh: A Differentiable Representation for General Meshes" (NeurIPS 2024)

vscode-mesh-viewer. A 3D mesh viewer for vscode

Objaverse-downloader. Instead of downloading the entire 8.9 TB dataset, download only files you need!

meshgpt-pytorch. Implementation of MeshGPT, SOTA Mesh generation using Attention, in Pytorch

elatentlpips. Author's Implementation for E-LatentLPIPS

GVHMR. Code for "GVHMR: World-Grounded Human Motion Recovery via Gravity-View Coordinates", Siggraph Asia 2024, TPAMI 2026

weights2weightspp. Exploring non-linear latent spaces with w2w

BlenderProc. A procedural Blender pipeline for photorealistic training image generation

4D-Humans. 4DHumans: Reconstructing and Tracking Humans with Transformers

PhysGaussian. [CVPR 2024 Highlight] PhysGaussian: Physics-Integrated 3D Gaussians for Generative Dynamics

gsplat. CUDA accelerated rasterization of gaussian splatting

PyTorch-VAE. A Collection of Variational Autoencoders (VAE) in PyTorch.

ctrl-x. Official implementation of "Ctrl-X: Controlling Structure and Appearance for Text-To-Image Generation Without Guidance" (NeurIPS 2024)

Awesome-Diffusion-Models. A collection of resources and papers on Diffusion Models

latentsplat. [ECCV 2024] Implementation of latentSplat: Autoencoding Variational Gaussians for Fast Generalizable 3D Reconstruction

Awesome-Controllable-Diffusion. Papers and resources on Controllable Generation using Diffusion Models, including ControlNet, DreamBooth, IP-Adapter.

Awesome-Controllable-T2I-Diffusion-Models. A collection of resources on controllable generation with text-to-image diffusion models.

mvdream_diffusers. A unified diffusers implementation for MVDream and ImageDream

LGM. [ECCV 2024 Oral] LGM: Large Multi-View Gaussian Model for High-Resolution 3D Content Creation.

dreamfusionacc. A minimal PyTorch implementation of text-to-3D models, e.g., DreamFusion and Magic3D, based on NerfAcc

MVDream-threestudio. 3D generation code for MVDream

IP-Adapter. The image prompt adapter is designed to enable a pretrained text-to-image diffusion model to generate images with image prompt.

Wonder3D. Single Image to 3D using Cross-Domain Diffusion for 3D Generation

MVDream. Multi-view Diffusion for 3D Generation

Tune-A-Video. [ICCV 2023] Tune-A-Video: One-Shot Tuning of Image Diffusion Models for Text-to-Video Generation

stable-dreamfusion. Text-to-3D & Image-to-3D & Mesh Exportation with NeRF + Diffusion.

CogVideo. text and image to video generation: CogVideoX (2024) and CogVideo (ICLR 2023)

academicpages.github.io. Github Pages template based upon HTML and Markdown for personal, portfolio-based websites.

jonbarron.github.io. HTML

awesome-3D-generation. A curated list of awesome 3d generation papers

Awesome-Video-Diffusion. A curated list of recent diffusion models for video generation, editing, and various other applications.

Gamba. [TPAMI] Gamba: Marry Gaussian Splatting with Mamba for single view 3D reconstruction

DiT-3D. 🔥🔥🔥Official Codebase of "DiT-3D: Exploring Plain Diffusion Transformers for 3D Shape Generation"

ZhiClass-PKU. CSS

Pointnet_Pointnet2_pytorch. PointNet and PointNet++ implemented by pytorch (pure python) and on ModelNet, ShapeNet and S3DIS.

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 👇

DreamPhysics. [AAAI 2025] DreamPhysics: Learning Physics-Based 3D Dynamics with Video Diffusion Priors

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

RAIN-GS. Code for "Relaxing Accurate Initialization Constraint for 3D Gaussian Splatting" by Jaewoo Jung, Jisang Han, Honggyu An, Jiwon Kang, Seonghoon Park, and Seungryong Kim

viser. Web-based 3D visualization in Python

semantic-gaussians. [TCSVT 2026] Official implemetation of the paper "Semantic Gaussians: Open-Vocabulary Scene Understanding with 3D Gaussian Splatting".

splat. WebGL 3D Gaussian Splat Viewer