Related papers: rd-spiral: An open-source Python library for learn…
In diffusion models, samples are generated through an iterative refinement process, requiring hundreds of sequential model evaluations. Several recent methods have introduced approximations (fewer discretization steps or distillation) to…
We develop numerical methods for reaction-diffusion systems based on the equations of fluctuating hydrodynamics (FHD). While the FHD formulation is formally described by stochastic partial differential equations (SPDEs), it becomes similar…
Powered by multimodal text-to-image priors, diffusion-based super-resolution excels at synthesizing intricate details; however, models trained on synthetic low-resolution (LR) and high-resolution (HR) image pairs often degrade when applied…
This study introduces amangkurat, an open-source Python library designed for the robust numerical simulation of relativistic scalar field dynamics governed by the nonlinear Klein-Gordon equation in $(1+1)$D spacetime. The software…
In this work animations of the random walk movement using a freeware Algodoo were done in order to support teaching the concepts of Brownian Motion. The random walk movement were simulate considering elastic collision between the particles…
The run-and-tumble (RT) dynamics followed by bacterial swimmers gives rise first to a ballistic motion due to their persistence, and later, through consecutive tumbles, to a diffusive process. Here we investigate how long it takes for a…
Pre-trained diffusion models have shown great potential in real-world image super-resolution (Real-ISR) tasks by enabling high-resolution reconstructions. While one-step diffusion (OSD) methods significantly improve efficiency compared to…
We have developed a multidimensional radiation hydrodynamics code to simulate the interaction of radial stellar pulsation and convection for full amplitude pulsating models. Convection is computed using large eddy simulations. Here we…
The present paper describes a combined experimental and high performance computing study of new specific behaviors of the Strato-Rotational Instability (SRI). The confrontation of low frequency Particle Image Velocimetry (PIV) experimental…
Reinforcement learning has emerged as a powerful tool for improving diffusion-based text-to-image models, but existing methods are largely limited to single-task optimization. Extending RL to multiple tasks is challenging: joint…
Constrained reinforcement learning (RL) seeks high-performance policies under safety constraints. We focus on an offline setting where the agent has only a fixed dataset -- common in realistic tasks to prevent unsafe exploration. To address…
This paper proposes Instruct 4D-to-4D that achieves 4D awareness and spatial-temporal consistency for 2D diffusion models to generate high-quality instruction-guided dynamic scene editing results. Traditional applications of 2D diffusion…
Recent literature has effectively leveraged diffusion models trained on continuous variables as priors for solving inverse problems. Notably, discrete diffusion models with discrete latent codes have shown strong performance, particularly…
Current 3D stylization methods often assume static scenes, which violates the dynamic nature of our real world. To address this limitation, we present S-DyRF, a reference-based spatio-temporal stylization method for dynamic neural radiance…
We introduce dro, an open-source Python library for distributionally robust optimization (DRO) for regression and classification problems. The library implements 14 DRO formulations and 9 backbone models, enabling 79 distinct DRO methods.…
We introduce an adaptable kinematic modelling tool called ROHSA-SNAPD, "Spatially Non-parametric Approach to PSF Deconvolution using ROHSA". ROHSA-SNAPD utilises kinematic regularisation to forward model the intrinsic emission-line flux and…
We are developing a 3D radiation hydrodynamics code to simulate the interaction of convection and pulsation in classical variable stars. One key goal is the ability to carry these simulations to full amplitude in order to compare them with…
Spontaneous pattern formation in homogeneous systems is ubiquitous in nature. Although Turing demonstrated that spatial patterns can emerge in reaction-diffusion (RD) systems when the homogeneous state becomes linearly unstable, it remains…
Two dimensional passive scalar turbulence is studied by means of a k-space diffusion model based on a third order differential approximation. This simple description of local nonlinear interactions in Fourier space is shown to present a…
The Rayleigh-Taylor (RT) instability is ubiquitously observed, yet has traditionally been studied using ideal fluid models. Collisionality can vary strongly across the fluid interface, and previous work demonstrates the necessity of kinetic…