Related papers: NoiseAR: AutoRegressing Initial Noise Prior for Di…

Recovering high-dimensional signals from corrupted measurements is a central challenge in inverse problems. Recent advances in generative diffusion models have shown remarkable empirical success in providing strong data-driven priors, but…

Although diffusion models with strong visual priors have emerged as powerful dense prediction backbones, they overlook a core limitation: the stochastic noise at the core of diffusion sampling is inherently misaligned with dense prediction…

Diffusion-based generative models (DBGMs) perturb data to a target noise distribution and reverse this process to generate samples. The choice of noising process, or inference diffusion process, affects both likelihoods and sample quality.…

The post-training pipeline for diffusion models currently has two stages: supervised fine-tuning (SFT) on curated data and reinforcement learning (RL) with reward models. A fundamental gap separates them. SFT optimizes the denoiser only on…

We propose an efficient framework for amortized conditional inference by leveraging exact conditional score-guided diffusion models to train a non-reversible neural network as a conditional generative model. Traditional normalizing flow…

Diffusion models have recently shown promise as powerful generative priors for inverse problems. However, conventional applications require solving the full reverse diffusion process and operating on noisy intermediate states, which poses…

Recent advances in video generation have been dominated by diffusion and flow-matching models, which produce high-quality results but remain computationally intensive and difficult to scale. In this work, we introduce VideoAR, the first…

Most recent real-world image super-resolution (Real-ISR) methods employ pre-trained text-to-image (T2I) diffusion models to synthesize the high-quality image either from random Gaussian noise, which yields realistic results but is slow due…

Denoising Diffusion models have shown remarkable performance in generating diverse, high quality images from text. Numerous techniques have been proposed on top of or in alignment with models like Stable Diffusion and Imagen that generate…

Most of the existing diffusion models use Gaussian noise for training and sampling across all time steps, which may not optimally account for the frequency contents reconstructed by the denoising network. Despite the diverse applications of…

Recommender systems often grapple with noisy implicit feedback. Most studies alleviate the noise issues from data cleaning perspective such as data resampling and reweighting, but they are constrained by heuristic assumptions. Another…

The strategy of combining diffusion-based generative models with classifiers continues to demonstrate state-of-the-art performance on adversarial robustness benchmarks. Known as adversarial purification, this exploits a diffusion model's…

Accurate noise modelling is important for training of deep learning reconstruction algorithms. While noise models are well known for traditional imaging techniques, the noise distribution of a novel sensor may be difficult to determine a…

Diffusion models have emerged as a powerful foundation model for visual generations. With an appropriate sampling process, it can effectively serve as a generative prior for solving general inverse problems. Current posterior sampling-based…

Super-resolution (SR) and image generation are important tasks in computer vision and are widely adopted in real-world applications. Most existing methods, however, generate images only at fixed-scale magnification and suffer from…

In supervised learning for image denoising, usually the paired clean images and noisy images are collected or synthesised to train a denoising model. L2 norm loss or other distance functions are used as the objective function for training.…

Accident anticipation is essential for proactive and safe autonomous driving, where even a brief advance warning can enable critical evasive actions. However, two key challenges hinder real-world deployment: (1) noisy or degraded sensory…

Diffusion models have achieved remarkable success across a wide range of generative tasks, yet their training paradigm largely treats injected noise as uniformly informative. In this work, we challenge this assumption and introduce…

Diffusion models have shown exceptional scaling properties in the image synthesis domain, and initial attempts have shown similar benefits for applying diffusion to unconditional text synthesis. Denoising diffusion models attempt to…

We present ControlSR, a new method that can tame Diffusion Models for consistent real-world image super-resolution (Real-ISR). Previous Real-ISR models mostly focus on how to activate more generative priors of text-to-image diffusion models…