pith. sign in

arxiv: 2310.12395 · v3 · pith:RFX3ZE35new · submitted 2023-10-19 · 💻 cs.LG · stat.ML

Closed-Form Diffusion Models

classification 💻 cs.LG stat.ML
keywords scoretrainingfunctionclosed-formerrorestimatormodelsneural
0
0 comments X
read the original abstract

Score-based generative models (SGMs) sample from a target distribution by iteratively transforming noise using the score function of the perturbed target. For any finite training set, this score function can be evaluated in closed form, but the resulting SGM memorizes its training data and does not generate novel samples. In practice, one approximates the score by training a neural network via score-matching. The error in this approximation promotes generalization, but neural SGMs are costly to train and sample, and the effective regularization this error provides is not well-understood theoretically. In this work, we instead explicitly smooth the closed-form score to obtain an SGM that generates novel samples without training. We analyze our model and propose an efficient nearest-neighbor-based estimator of its score function. Using this estimator, our method achieves competitive sampling times while running on consumer-grade CPUs.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 8 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Training-Free Generative Sampling via Moment-Matched Score Smoothing

    stat.ML 2026-05 unverdicted novelty 7.0

    MM-SOLD is a training-free particle sampler whose large-particle limit converges to a moment-matched Gibbs distribution obtained by exponentially tilting a score-smoothed target.

  2. Is Flow Matching Just Trajectory Replay for Sequential Data?

    stat.ML 2026-02 unverdicted novelty 7.0

    Flow matching on time series targets a closed-form nonparametric velocity field that is a similarity-weighted mixture of observed transition velocities, making neural models approximations to an ideal memory-augmented...

  3. On The Hidden Biases of Flow Matching Samplers

    stat.ML 2025-12 unverdicted novelty 7.0

    Empirical flow matching introduces coupled biases from plug-in estimation, including altered statistical targets, non-gradient minimizers, and non-unique dynamics via flux-null fields, with base distribution controlli...

  4. Substitution-Based Analysis of Structural Novelty for Generative Models of Materials

    cs.LG 2026-06 unverdicted novelty 6.0

    81-92% of chemically valid and metastable crystals from generative models are training duplicates or substitution-derived, with low-symmetry cases showing interpolation and high-symmetry cases showing memorization.

  5. Path convergence in diffusion models

    cond-mat.stat-mech 2026-06 unverdicted novelty 6.0

    In a 1D test case, backward diffusion paths converge to the infinite-pattern limit on a 1/sqrt(p) scale with infinite mean square deviation, enabling an extrapolation algorithm for density estimation and generalization.

  6. The two clocks and the innovation window: When and how generative models learn rules

    cs.LG 2026-05 unverdicted novelty 6.0

    Generative models learn rules before memorizing data, creating an innovation window whose width depends on dataset size and rule complexity, observed in both diffusion and autoregressive architectures.

  7. Diffusion Models Memorize in Training -- and Generalize in Inference

    cs.LG 2026-03 unverdicted novelty 6.0

    Diffusion models overfit denoising loss at intermediate noise but generalize in inference as model error smooths the flow field and sampling paths avoid memorized noisy training data.

  8. Demystifying Transition Matching: When and Why It Can Beat Flow Matching

    cs.LG 2025-10 unverdicted novelty 6.0

    TM outperforms FM for well-separated modes with non-negligible variance by preserving covariance via stochastic latent updates, with the gap closing as variance approaches zero.