English
Related papers

Related papers: Diffusion: Quiescence and Perturbation

200 papers

The increased model capacity of Diffusion Transformers (DiTs) and the demand for generating higher resolutions of images and videos have led to a significant rise in inference latency, impacting real-time performance adversely. While prior…

Computer Vision and Pattern Recognition · Computer Science 2024-11-22 Xibo Sun , Jiarui Fang , Aoyu Li , Jinzhe Pan

Rendering highly scattering participating media using brute force path tracing is a challenge. The diffusion approximation reduces the problem to solving a simple linear partial differential equation. Flux-limited diffusion introduces…

Graphics · Computer Science 2018-07-03 David Koerner , Jamie Portsmouth , Wenzel Jakob

The aim of this paper is to discuss the constructivity of the method originally introduced by U. Bessi to approach the phenomenon of topological instability commonly known as Arnold's Diffusion. By adapting results and proofs from existing…

Dynamical Systems · Mathematics 2023-06-23 Alessandro Fortunati

The general theoretical description of spin self-diffusion under nonlinear gradient is proposed, which extends the effective phase diffusion method for linear gradient field. Based on the phase diffusion, the proposed method reveals the…

Chemical Physics · Physics 2024-05-30 Guoxing Lin

We consider a stochastically perturbed reaction diffusion equation in a bounded interval, with boundary conditions imposing the two stable phases at the endpoints. We investigate the asymptotic behavior of the front separating the two…

Mathematical Physics · Physics 2016-02-11 L. Bertini , S. Brassesco , P. Buttà

Two years ago, Stable Diffusion achieved super-human performance at generating images with super-human numbers of fingers. Following the steady decline of its technical novelty, we propose Stale Diffusion, a method that solidifies and…

Computer Vision and Pattern Recognition · Computer Science 2024-04-02 Joao F. Henriques , Dylan Campbell , Tengda Han

We study reaction-diffusion systems where diffusion is by jumps whose sizes are distributed exponentially. We first study the Fisher-like problem of propagation of a front into an unstable state, as typified by the A+B $\to$ 2A reaction. We…

Statistical Mechanics · Physics 2009-11-11 Elisheva Cohen , David A. Kessler

Diffusion models have become leading approaches for high-fidelity image generation. Recent DiT-based diffusion models, in particular, achieve strong prompt adherence while producing high-quality samples. We propose SHIFT, a simple but…

Computer Vision and Pattern Recognition · Computer Science 2026-04-13 Nina Konovalova , Andrey Kuznetsov , Aibek Alanov

We study quantum walks on general graphs from the point of view of scattering theory. For a general finite graph we choose two vertices and attach one half line to each. We are interested in walks that proceed from one half line, through…

Quantum Physics · Physics 2009-11-10 Edgar Feldman , Mark Hillery

We study a quantum particle propagating through a ``quantum mechanically chaotic'' background, described by parametric random matrices with only short range spatial correlations. The particle is found to exhibit turbulent-like diffusion…

chao-dyn · Physics 2009-10-30 Dimitri Kusnezov , Aurel Bulgac , Gui Do Dang

Conventional approaches for simulating steady-state distributions of particles under diffusive and advective transport at high P\'eclet numbers involve solving the diffusion and advection equations in at least two dimensions. Here, we…

We consider a singularly perturbed convection-diffusion problem that has in addition a shift term. We show a solution decomposition using asymptotic expansions and a stability result. Based upon this we provide a numerical analysis of high…

Numerical Analysis · Mathematics 2022-07-20 Mirjana Brdar , Sebastian Franz , Lars Ludwig , Hans-Görg Roos

There is a bias in the inference pipeline of most diffusion models. This bias arises from a signal leak whose distribution deviates from the noise distribution, creating a discrepancy between training and inference processes. We demonstrate…

Computer Vision and Pattern Recognition · Computer Science 2023-10-25 Martin Nicolas Everaert , Athanasios Fitsios , Marco Bocchio , Sami Arpa , Sabine Süsstrunk , Radhakrishna Achanta

Diffusion-based image compression methods have achieved notable progress, delivering high perceptual quality at low bitrates. However, their practical deployment is hindered by significant inference latency and heavy computational overhead,…

Computer Vision and Pattern Recognition · Computer Science 2026-02-03 Yiwen Jia , Hao Wei , Yanhui Zhou , Chenyang Ge

The analysis of diffusion processes in real-world propagation scenarios often involves estimating variables that are not directly observed. These hidden variables include parental relationships, the strengths of connections between nodes,…

Social and Information Networks · Computer Science 2016-05-12 Shohreh Shaghaghian , Mark Coates

We consider the general problem of the first passage distribution of particles whose displacements are subject to time delays. We show that this problem gives rise to a \emph{propagation-dispersion equation} which is obtained as the…

Statistical Mechanics · Physics 2009-11-10 Jean Pierre Boon , Patrick Grosfils , James F. Lutsko

Macro placement is a vital step in digital circuit design that defines the physical location of large collections of components, known as macros, on a 2D chip. Because key performance metrics of the chip are determined by the placement,…

Machine Learning · Computer Science 2025-06-11 Vint Lee , Minh Nguyen , Leena Elzeiny , Chun Deng , Pieter Abbeel , John Wawrzynek

The subcritical transition to turbulence, as occurs in pipe flow, is believed to generically be a phase transition in the directed percolation universality class. At its heart is a balance between the decay rate and proliferation rate of…

Fluid Dynamics · Physics 2022-09-16 Anna Frishman , Tobias Grafke

This paper presents Diffusion Forcing, a new training paradigm where a diffusion model is trained to denoise a set of tokens with independent per-token noise levels. We apply Diffusion Forcing to sequence generative modeling by training a…

Machine Learning · Computer Science 2024-12-11 Boyuan Chen , Diego Marti Monso , Yilun Du , Max Simchowitz , Russ Tedrake , Vincent Sitzmann

Divergence and vorticity damping, which operate upon horizontal divergence and relative vorticity, are explicit diffusion mechanisms used in dynamical cores to ensure stability. To avoid numerical blow-up from excessively strong diffusion,…

Numerical Analysis · Mathematics 2026-02-19 Timothy C. Andrews , Christiane Jablonowski