Fast and Scalable Analytical Diffusion

• URL: http://arxiv.org/abs/2602.16498v1
• Date: Wed, 18 Feb 2026 14:41:09 GMT
• Title: Fast and Scalable Analytical Diffusion
• Authors: Xinyi Shang, Peng Sun, Jingyu Lin, Zhiqiang Shen,
• Abstract summary: Analytical diffusion models offer a mathematically transparent path to generative modeling by formulating the denoising score as an empirical-Bayes posterior mean.<n>Standard formulation requires a full-dataset scan at every timestep, scaling linearly with dataset size.<n>We propose Dynamic Time-Aware Golden Subset Diffusion (GoldDiff), a training-free framework that decouples inference from dataset size.
• Score: 43.4234198410875
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Analytical diffusion models offer a mathematically transparent path to generative modeling by formulating the denoising score as an empirical-Bayes posterior mean. However, this interpretability comes at a prohibitive cost: the standard formulation necessitates a full-dataset scan at every timestep, scaling linearly with dataset size. In this work, we present the first systematic study addressing this scalability bottleneck. We challenge the prevailing assumption that the entire training data is necessary, uncovering the phenomenon of Posterior Progressive Concentration: the effective golden support of the denoising score is not static but shrinks asymptotically from the global manifold to a local neighborhood as the signal-to-noise ratio increases. Capitalizing on this, we propose Dynamic Time-Aware Golden Subset Diffusion (GoldDiff), a training-free framework that decouples inference complexity from dataset size. Instead of static retrieval, GoldDiff uses a coarse-to-fine mechanism to dynamically pinpoint the ''Golden Subset'' for inference. Theoretically, we derive rigorous bounds guaranteeing that our sparse approximation converges to the exact score. Empirically, GoldDiff achieves a $\bf 71 \times$ speedup on AFHQ while matching or achieving even better performance than full-scan baselines. Most notably, we demonstrate the first successful scaling of analytical diffusion to ImageNet-1K, unlocking a scalable, training-free paradigm for large-scale generative modeling.

Related papers

• Efficient Sampling with Discrete Diffusion Models: Sharp and Adaptive Guarantees [9.180350432640912]
  We study the sampling efficiency of score-based discrete diffusion models under a continuous-time Markov chain (CTMC) formulation.<n>For uniform discrete diffusion, we show that the $$-leaping algorithm achieves an complexity of order $tilde O(d/varepsilon)$.<n>For masking discrete diffusion, we introduce a modified $$-leaping sampler whose convergence rate is governed by an intrinsic information-theoretic quantity.
  arXiv  Detail & Related papers  (2026-02-16T18:48:17Z)
• Sublinear iterations can suffice even for DDPMs [20.394390032998036]
  We show that a "shifted composition rule" algorithm enjoys favorable discretization properties under appropriate assumptions.<n>This is the first sublinear complexity bound for pure DDPM sampling.<n>We also provide experimental validation of the advantages of our method, showing that it performs well in practice with pre-trained image synthesis models.
  arXiv  Detail & Related papers  (2025-11-06T22:15:19Z)
• Noise Hypernetworks: Amortizing Test-Time Compute in Diffusion Models [57.49136894315871]
  New paradigm of test-time scaling has yielded remarkable breakthroughs in reasoning models and generative vision models.<n>We propose one solution to the problem of integrating test-time scaling knowledge into a model during post-training.<n>We replace reward guided test-time noise optimization in diffusion models with a Noise Hypernetwork that modulates initial input noise.
  arXiv  Detail & Related papers  (2025-08-13T17:33:37Z)
• Inference-Time Scaling of Diffusion Language Models with Particle Gibbs Sampling [70.8832906871441]
  We study how to steer generation toward desired rewards without retraining the models.<n>Prior methods typically resample or filter within a single denoising trajectory, optimizing rewards step-by-step without trajectory-level refinement.<n>We introduce particle Gibbs sampling for diffusion language models (PG-DLM), a novel inference-time algorithm enabling trajectory-level refinement while preserving generation perplexity.
  arXiv  Detail & Related papers  (2025-07-11T08:00:47Z)
• Faster Diffusion Models via Higher-Order Approximation [28.824924809206255]
  We propose a principled, training-free sampling algorithm that requires only the order of $$ d1+2/K varepsilon-1/K $$ score function evaluations.<n>Our theory is robust vis-a-vis inexact score estimation, degrading gracefully as the score estimation error increases.<n>More broadly, our work develops a theoretical framework towards understanding the efficacy of high-order methods for accelerated sampling.
  arXiv  Detail & Related papers  (2025-06-30T16:49:03Z)
• Posterior Sampling with Denoising Oracles via Tilted Transport [37.14320147233444]
  We introduce the textittilted transport technique, which leverages the quadratic structure of the log-likelihood in linear inverse problems. We quantify the conditions under which this boosted posterior is strongly log-concave, highlighting the dependencies on the condition number of the measurement matrix. The resulting posterior sampling scheme is shown to reach the computational threshold predicted for sampling Ising models.
  arXiv  Detail & Related papers  (2024-06-30T16:11:42Z)
• Provably Robust Score-Based Diffusion Posterior Sampling for Plug-and-Play Image Reconstruction [31.503662384666274]
  In science and engineering, the goal is to infer an unknown image from a small number of measurements collected from a known forward model describing certain imaging modality. Motivated Score-based diffusion models, due to its empirical success, have emerged as an impressive candidate of an exemplary prior in image reconstruction.
  arXiv  Detail & Related papers  (2024-03-25T15:58:26Z)
• Adaptive Federated Learning Over the Air [108.62635460744109]
  We propose a federated version of adaptive gradient methods, particularly AdaGrad and Adam, within the framework of over-the-air model training. Our analysis shows that the AdaGrad-based training algorithm converges to a stationary point at the rate of $mathcalO( ln(T) / T 1 - frac1alpha ).
  arXiv  Detail & Related papers  (2024-03-11T09:10:37Z)
• Sparse Training of Discrete Diffusion Models for Graph Generation [45.103518022696996]
  We introduce SparseDiff, a novel diffusion model based on the observation that almost all large graphs are sparse. By selecting a subset of edges, SparseDiff effectively leverages sparse graph representations both during the noising process and within the denoising network. Our model demonstrates state-of-the-art performance across multiple metrics on both small and large datasets.
  arXiv  Detail & Related papers  (2023-11-03T16:50:26Z)
• Hierarchical Integration Diffusion Model for Realistic Image Deblurring [71.76410266003917]
  Diffusion models (DMs) have been introduced in image deblurring and exhibited promising performance. We propose the Hierarchical Integration Diffusion Model (HI-Diff), for realistic image deblurring. Experiments on synthetic and real-world blur datasets demonstrate that our HI-Diff outperforms state-of-the-art methods.
  arXiv  Detail & Related papers  (2023-05-22T12:18:20Z)
• ChiroDiff: Modelling chirographic data with Diffusion Models [132.5223191478268]
  We introduce a powerful model-class namely "Denoising Diffusion Probabilistic Models" or DDPMs for chirographic data. Our model named "ChiroDiff", being non-autoregressive, learns to capture holistic concepts and therefore remains resilient to higher temporal sampling rate.
  arXiv  Detail & Related papers  (2023-04-07T15:17:48Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.