Diffusion Model-Driven Spectral Noise Modeling and Generative Reconstruction Strategy.

Enhancing generalization in non-uniformly sampled NMR spectra reconstruction via accelerated conditional diffusion models

An accelerated conditional diffusion model for multidimensional NMR spectra reconstruction is presented, formulating the task as a probabilistic iterative denoising process that progressively refines undersampled spectra under physical constraints and outperforms both traditional and end-to-end deep learning algorithms...

Diffusion models in deep generative learning: a unified review of theories, architectures, and applications

STELT: A spatiotemporal deep learning framework for diffusion-ordered NMR spectroscopy reconstruction with artifact suppression.

STELT (Spatiotemporal Extraction Laplace Transform), a lightweight deep-learning framework based on spatiotemporal feature extraction, is proposed, which achieves superior reconstruction accuracy and noise suppression with significantly reduced computational overhead.

A Physics-Informed Multi-Scale Diffusion Model Framework for Physical Field Imputation from Randomly Missing Data

Non-intrusive optical measurement techniques are widely used to obtain high-resolution pressure, temperature, and velocity fields, but they often suffer from random data loss caused by geometric obstruction, surface reflection, or illumination non-uniformity. Conventional reconstruction methods usually depend on high...

Adaptive Momentum Benefits Score-Based Generative Models

Score-based Generative Models (SGMs) have achieved remarkable success in generation tasks by establishing a diffusion process that gradually perturbs real data into Gaussian noise and then learning a reverse process to reconstruct the data from noise. However, conventional denoising methods, such as Langevin dynamics a...

Residual Diffusion Implicit Models

Diffusion models achieve state-of-the-art results across multiple tasks. However, in inverse problems, standard initialization from pure Gaussian noise misaligns the generative process with real-world degradations. More recent methods such as diffusion bridges impose strict endpoint constraints and often require long r...