Diffusion models research tackles outliers, efficiency, and theory · 10 sources tracked
ByPulseAugur Editorial·[15 sources]·
Recent research explores advancements in diffusion models, focusing on improving their robustness, efficiency, and theoretical understanding. Papers address challenges like outlier data in inverse problems, scaling reinforcement learning for preference tuning, and developing theoretical frameworks for speciation and generalization. Additionally, new methods are proposed for efficient inference in diffusion transformers and for pixel-space diffusion models, aiming to enhance training stability and generation quality.
AI
IMPACT
Advances in diffusion model robustness, efficiency, and theoretical understanding could accelerate their adoption in various applications.
RANK_REASON
Multiple arXiv papers published on diffusion models, covering theoretical and applied aspects.
arXiv:2406.06837v2 Announce Type: cross Abstract: A Bayesian data assimilation scheme is formulated for advection-dominated advective and diffusive evolutionary problems, based upon the Dynamic Likelihood (DLF) approach to filtering. The DLF was developed specifically for hyperbo…
arXiv:2605.09477v2 Announce Type: replace-cross Abstract: Methods based on diffusion models (DMs) for solving inverse problems (IPs) have recently achieved remarkable performance. However, DM-based methods typically struggle against outliers, which are common in real-world measur…
arXiv:2608.23798v1 Announce Type: new Abstract: Speciation in generative diffusion models denotes the emergence of distinct stable branches during denoising, through which initially undifferentiated trajectories progressively commit to different data classes. In this work we deve…
arXiv:2608.23664v1 Announce Type: cross Abstract: Reward fine-tuning is becoming an important tool for adapting diffusion models to human preferences and task-specific objectives, but existing methods largely inherit policy-gradient machinery from large language models. Unlike au…
Diffusion models and flow-based models have recently become the dominant paradigms in generative modeling, largely due to their ability to learn rich, multi-level visual representations through large-scale training. This creates a bidirectional relationship between generative mod…
arXiv:2608.23429v1 Announce Type: cross Abstract: Diffusion Transformers (DiTs) have shown strong performance in high-fidelity image generation, but their sampling process remains computationally intensive due to full model execution at every timestep. While cache-based accelerat…
arXiv:2601.21242v2 Announce Type: replace-cross Abstract: Motivated by challenges in conditional generative modeling, where the target conditional density takes the form of a ratio f1 over f2, this paper develops a theoretical framework for approximating such ratio-type functiona…
arXiv:2608.27080v1 Announce Type: new Abstract: Many scientific and engineering applications require estimating unknown parameters from experimentally observable data -- an inverse problem that is inherently challenging due to nonlinearity, noise, and ill-posedness. In this paper…
arXiv:2608.25998v1 Announce Type: new Abstract: The perception-distortion trade-off poses a fundamental challenge in single-image super-resolution (SR). Although diffusion-based SR methods excel at generating perceptually realistic images, achieving high fidelity remains a key li…
arXiv stat.ML
TIER_1English(EN)·Hugo Latourelle-Vigeant, Sinho Chewi, Aram-Alexandre Pooladian, John Sous, Theodor Misiakiewicz·
arXiv:2608.23938v1 Announce Type: new Abstract: Modern score-based generative models have achieved remarkable empirical success in high-dimensional tasks such as image, audio, and video synthesis. These models reduce distribution learning to a sequence of regression problems that…
arXiv:2608.24068v1 Announce Type: new Abstract: Diffusion models and flow-based models have recently become the dominant paradigms in generative modeling, largely due to their ability to learn rich, multi-level visual representations through large-scale training. This creates a b…
arXiv:2608.23554v1 Announce Type: cross Abstract: Discrete diffusion models offer a promising alternative to autoregressive generation by enabling parallel updates, but their sampling efficiency can depend strongly on the choice of the forward process and the sampler. For the uni…
arXiv:2509.04719v3 Announce Type: replace-cross Abstract: The widespread adoption of diffusion models for image generation necessitates efficient parallel inference to manage their substantial computational overhead. However, current parallel inference paradigms primarily target …
arXiv:2608.21972v1 Announce Type: cross Abstract: The sampling process of Denoising Diffusion Probabilistic Models (DDPMs) can be accelerated by leveraging second-order information in the form of approximations to the denoising posterior covariance -- allowing samples of acceptab…
arXiv:2608.21885v1 Announce Type: new Abstract: Pixel-space diffusion models directly model image distributions but remain difficult to optimize. Recent methods alleviate this challenge through target reparameterization, while still relying on a fixed clean-image target throughou…
