Highlights
- Pro
Stars
[ICML 2025] M2PDE: Compositional Generative Multiphysics and Multi-component PDE Simulation.
[ICLR'25] Artificial Kuramoto Oscillatory Neurons
[ICLR 2025] CL-DiffPhyCon achieves closed-loop diffusion control of physical systems with significant speedup of sampling efficiency
Official PyTorch implementation for "Large Language Diffusion Models"
[ICML 2025] The official implementation of the paper "On the Guidance of Flow Matching"
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.
A 15TB Collection of Physics Simulation Datasets
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.
[ICLR2025] Wavelet Diffusion Neural Operator (WDNO) uses diffusion models on wavelet space for generative PDE simulation and control.
[NeurIPS2024] DiffPhyCon uses generative models to control complex physical systems
The reinforcement learning training code for AgiBot X1.
[ICLR 2025] From anything to mesh like human artists. Official impl. of "MeshAnything: Artist-Created Mesh Generation with Autoregressive Transformers"
Official PyTorch implementation of Learning to (Learn at Test Time): RNNs with Expressive Hidden States
Author's Pytorch implementation of ICLR2023 paper Behavior Proximal Policy Optimization (BPPO).
TorchCFM: a Conditional Flow Matching library
[AAAI24] LE-PDE-UQ endows deep learning-based surrogate models with robust and efficient uncertainty quantification capabilities for both forward and inverse problems.
[ICLR24] A boundary-embedded neural operator that incorporates complex boundary shape and inhomogeneous boundary values
[ICLR24] CinDM uses compositional generative models to design boun 4D5E daries and initial states significantly more complex than the ones seen in training for physical simulation
Imitation learning algorithms with Co-training for Mobile ALOHA: ACT, Diffusion Policy, VINN
ViRel: Unsupervised Visual Relations Discovery with Graph-level Analogy
[ICLR23] First deep learning-based surrogate model that jointly learns the evolution model and optimizes computational cost via remeshing