Mengning Wu

JustGRPO. [ICML 2026 Outstanding Paper] Minimalist RL for Diffusion LLMs. 89.1% on GSM8K.

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

mira. Code for MIRA: Multiplayer Interactive World Models with Representation Autoencoders

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

opal-bridge. Cross-agent session translator (Claude Code <-> Codex CLI)

andrej-karpathy-skills. A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.

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

OpenMythos. A theoretical reconstruction of the Claude Mythos architecture, built from first principles using the available research literature.

flm. Official Codebase For paper "One-step Language Modeling via Continuous Denoising"

Sim2Reason. Sim2Reason: Solving Physics Olympiad via Reinforcement Learning on Physics Simulators. We present a method for turning physics simulators into scalable generators of question–answer pairs for improving LLM physical reasoning.

Awesome-Latent-Space. A paper list of Awesome Latent Space.

skills. Public repository for Agent Skills

nepa. PyTorch implementation of NEPA

Making-rFID-Predictive-of-Diffusion-gFID. Predicting the generation FID of latent diffusion, with a variant of reconstruction FID of Variational Auto-encoder.

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

calm. Official implementation of "Continuous Autoregressive Language Models"

BFM-Zero. BFM_Zero: A Promptable Behavioral Foundation Model for Humanoid Control Using Unsupervised Reinforcement Learning

OLMo-core. PyTorch building blocks for the OLMo ecosystem

gated_attention. The official implementation for [NeurIPS2025 Oral] Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free

diffusion-data-constraint. Official PyTorch implementation and models for paper "Diffusion Beats Autoregressive in Data-Constrained Settings". We find diffusion models are significantly more data-efficient than standard left to right autoregressive models, due to their ability to learn from different token orderings.

JiT. PyTorch implementation of JiT https://arxiv.org/abs/2511.13720

dLLM-RL. [ICLR 2026] Official code for TraceRL: Revolutionizing post-training for Diffusion LLMs, powering the SOTA TraDo series.

dllm. dLLM: Simple Diffusion Language Modeling

open-thoughts. Fully open data curation for reasoning models

diffusion-gpt. From baby GPT to diffusion GPT: An annotated implementation of a character-level discrete diffusion model (adapted from Karpathy’s baby GPT).

RLP. [ICLR 2026] Official PyTorch Implementation of RLP: Reinforcement as a Pretraining Objective

EBT. PyTorch Code for Energy-Based Transformers paper -- generalizable reasoning and scalable learning

llm-liar. HTML

AI-Scientist-v2. The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search

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

AI-Scientist-ICLR2025-Workshop-Experiment. Python

mle-bench. MLE-bench is a benchmark for measuring how well AI agents perform at machine learning engineering

mem0. Universal memory layer for AI Agents

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

dinov3. Reference PyTorch implementation and models for DINOv3

mirage. Mirage Persistent Kernel: Compiling LLMs into a MegaKernel

happy-llm. 📚 从零开始构建大模型

limit-of-RLVR. repo for paper https://arxiv.org/abs/2504.13837

Spurious_Rewards. Python

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

DiffuLLaMA. [ICLR2025] DiffuGPT and DiffuLLaMA: Scaling Diffusion Language Models via Adaptation from Autoregressive Models

shit.

chain-of-thought-hub. Benchmarking large language models' complex reasoning ability with chain-of-thought prompting

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

open-r1. Fully open reproduction of DeepSeek-R1

LLaDA. Official PyTorch implementation for "Large Language Diffusion Models"

RADD. Official PyTorch implementation for "Your Absorbing Discrete Diffusion Secretly Models the Conditional Distributions of Clean Data" (ICLR 2025)

any4lerobot. 🎁 A collection of utilities for LeRobot.

CFG-Zero-star. Official repo for CFG-Zero*

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

3D-Diffusion-Policy. [RSS 2024] 3D Diffusion Policy: Generalizable Visuomotor Policy Learning via Simple 3D Representations

Perturb-and-Revise. Official implementation of the paper "Perturb-and-Revise: Flexible 3D Editing with Generative Trajectories" (CVPR`25)

Awesome-ScalingLaws. A curated list of awesome resources dedicated to Scaling Laws for LLMs

zotero-pdf-translate. Translate PDF, EPub, webpage, metadata, annotations, notes to the target language. Support 20+ translate services.

zotero-better-notes. Everything about note management. All in Zotero.

repulsive-shells. C++

awesome-nerf-editing. 🧙🏻‍♂️A list of papers curated for you to dive into the Awesome Radiance Field-based 3D Editing.

uv. An extremely fast Python package and project manager, written in Rust.

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

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

LLaVA. [NeurIPS'23 Oral] Visual Instruction Tuning (LLaVA) built towards GPT-4V level capabilities and beyond.

Golden-Noise-for-Diffusion-Models. [ICCV2025] The code of our work "Golden Noise for Diffusion Models: A Learning Framework".

Awesome-Image-Editing. A Survey of Image Editing

awesome-llm-interpretability. A curated list of Large Language Model (LLM) Interpretability resources.

LIBERO. Benchmarking Knowledge Transfer in Lifelong Robot Learning

CFGpp. Official repository for "CFG++: manifold-constrained classifier free guidance for diffusion models" (ICLR2025)

phd-cs-us. A open-source guide that demystifies how U.S. universities evaluate and admit students into Computer Science PhD programs.

mdlm. [NeurIPS 2024] Simple and Effective Masked Diffusion Language Model

openpi. Python

lm-evaluation-harness. A framework for few-shot evaluation of language models.

llama-cookbook. Welcome to the Llama Cookbook! This is your go to guide for Building with Llama: Getting started with Inference, Fine-Tuning, RAG. We also show you how to solve end to end problems using Llama model family and using them on various provider services

lmms-eval. One-for-All Multimodal Evaluation Toolkit Across Text, Image, Video, and Audio Tasks

SAELens. Training Sparse Autoencoders on Language Models

ml-depth-pro. Depth Pro: Sharp Monocular Metric Depth in Less Than a Second.

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

diffae. Official implementation of Diffusion Autoencoders

robotics-toolbox-python. Robotics Toolbox for Python

puma560_description. ROS package for PUMA560 urdf model

DataSciencePsychNeuro. Repository for the Data Science for Psychology and Neuroscience course at CMU (Vargas/Verstynen)

taming-transformers. Taming Transformers for High-Resolution Image Synthesis

vqvae. A pytorch implementation of the vector quantized variational autoencoder (https://arxiv.org/abs/1711.00937)

conditional-flow-matching. TorchCFM: a Conditional Flow Matching library

ScaMo_code. Python

muse-maskgit-pytorch. Implementation of Muse: Text-to-Image Generation via Masked Generative Transformers, in Pytorch

MUSE. A library for Multilingual Unsupervised or Supervised word Embeddings

maskgit. Official Jax Implementation of MaskGIT

SMDM. Official PyTorch implementation for ICLR2025 paper "Scaling up Masked Diffusion Models on Text"

datablations. Scaling Data-Constrained Language Models

FlowEdit. Official implementation of the paper: "FlowEdit: Inversion-Free Text-Based Editing Using Pre-Trained Flow Models"

Rectified-Diffusion. [ICLR 2025] Rectified Diffusion: Straightness Is Not Your Need

CrossFlow. [CVPR2025] PyTorch-based reimplementation of CrossFlow, as proposed in 'Flowing from Words to Pixels: A Noise-Free Framework for Cross-Modality Evolution'

score_sde_pytorch. PyTorch implementation for Score-Based Generative Modeling through Stochastic Differential Equations (ICLR 2021, Oral)

score_sde. Official code for Score-Based Generative Modeling through Stochastic Differential Equations (ICLR 2021, Oral)

pix2pix. Image-to-image translation with conditional adversarial nets

OmniEdit. Official Repo for Paper "OmniEdit: Building Image Editing Generalist Models Through Specialist Supervision" [ICLR2025]

HQ-Edit. [ICLR 2025] HQ-Edit: A High-Quality and High-Coverage Dataset for General Image Editing

Cosmos-Tokenizer. A suite of image and video neural tokenizers

flow_matching. A PyTorch library for implementing flow matching algorithms, featuring continuous and discrete flow matching implementations. It includes practical examples for both text and image modalities.

stochastic-interpolants. Jupyter Notebook

mnist-dits. Diffusion Transformers (DiTs) trained on MNIST dataset