[ICLR 2025] CL-DiffPhyCon achieves closed-loop diffusion control of physical systems with significant speedup of sampling efficiency

This repo is the code for T-SCEND, a novel framework that significantly improves diffusion model’s reasoning capabilities with better energy-based training and scaling up test-time computation.

Physical Symbolic Optimization

High-Performance Symbolic Regression in Python and Julia

This repository is a collection of papers, methods, benchmarks and datasets of SR(Symbolic Regression).

The official implementation of the paper "On the Guidance of Flow Matching"

Official codebase for Decision Transformer: Reinforcement Learning via Sequence Modeling.

Official codebase for Exact Energy-Guided Diffusion Sampling via Contrastive Energy Prediction （ICML 2023）

Code for the paper "Planning with Diffusion for Flexible Behavior Synthesis"

[NeurIPS 2024] Official implementation of NeurIPS 2024 paepr "Flow Priors for Linear Inverse Problems via Iterative Corrupted Trajectory Matching"

Official Implementation of Rectified Flow (ICLR2023 Spotlight)

Code for PnP-Flow: Plug-and-Play image restoration with Flow Matching (ICLR 2025)

Tokamak plasma operation control through multi-objective reinforcement learning in KSTAR

Stellarator Equilibrium and Optimization Suite

A tokamak (nuclear fusion reactor) simulator with LSTM-based neural network (KSTAR-NN)

The Python scripts for training the ML models for tearing prediction & avoidance

AI design of tokamak operation for autonomous control of fusion plasma.

Grad-Shafranov equation solver for hydrostatic equilibrium in Tokamak plasma

TORAX: Tokamak transport simulation in JAX

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.

This is project page for the paper "RG-Flow: a hierarchical and explainable flow model based on renormalization group and sparse prior". Paper link: https://arxiv.org/abs/2010.00029

CleanDiffuser: An Easy-to-use Modularized Library for Diffusion Models in Decision Making

PyTorch implementation of Soft Actor-Critic (SAC), Twin Delayed DDPG (TD3), Actor-Critic (AC/A2C), Proximal Policy Optimization (PPO), QT-Opt, PointNet..

AlphaFold 3 inference pipeline.

Official codebase for Exact Energy-Guided Diffusion Sampling via Contrastive Energy Prediction

[NeurIPS2024] DiffPhyCon uses generative models to control complex physical systems

Differentiable ODE solvers with full GPU support and O(1)-memory backpropagation.