Pith. sign in

REVIEW 2 major objections 4 minor 2 cited by

Adaptivity and Convergence of Probability Flow ODEs in Diffusion Generative Models

T0 review · 2 major / 4 minor · reviewed 2026-08-09 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The paper proves that, with the DDIM coefficient choice, the probability flow ODE sampler converges in total variation at rate $O((k+\log d)\log^3 T/T)$ plus score-error terms, where $k$ is the intrinsic dimension of the data support.

desk verdict A plausible new O(k/T) bound for the probability flow ODE, but the advertised 'sampler' claim outruns the theorem, which stops at Y1; the k-scaling rests on an imported unproved lemma. read the letter →

arxiv 2501.18863 v1 pith:LRR7GI6N submitted 2025-01-31 stat.ML cs.LG

Jiaqi Tang , Yuling Yan This is my paper · ORCID
classification stat.MLcs.LG
keywords score-basedgenerativemodelsprobabilityflowODEDDIMcoefficientsintrinsicdimensionlow-dimensionalstructuretotalvariationconvergenceadaptivesamplingratesdiffusion
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Score-based generative models typically turn noise into data by reversing a diffusion, and the deterministic probability flow ODE is the version used most in practice. This paper proves that this sampler adapts to low-dimensional structure: with the DDIM coefficient choice $\eta^\star_t$, the total variation distance between the forward marginal and the sampled marginal is bounded by $O((k+\log d)\log^3 T/T)$ plus score-error terms, where $k$ is the intrinsic dimension of the data support measured by metric entropy. Under perfect score estimation the rate becomes $O(k/T)$, replacing the ambient dimension $d$ in the previous best $O(d/T)$ bound. Because natural images are believed to concentrate near low-dimensional manifolds, a $k/T$ rate is the theoretical counterpart of the practical speed of deterministic samplers.

What carries the argument

The engine of the proof is the deterministic one-step map $\phi_t(x)=x+\eta^\star_t s_t(x)$ analyzed on high-probability tubes $\mathcal{T}_t$ built from an $\varepsilon$-net of the support. The key comparison Lemma 3 bounds the density ratio $p_{\sqrt{\alpha_t}X_{t-1}}(\phi_t(x))/p_{X_t}(x)$ divided by $p_{\phi_t(Y_t)}(\phi_t(x))/p_{Y_t}(x)$, showing it is exponentially close to $1$ with exponent $k\log^3 T/T^2$ plus score and Jacobian errors; summing these factors over $T$ steps yields the $k/T$ dependence. The intrinsic dimension enters when bounding the average conditional covariance $\Sigma_{\alpha_t}(X_t)$ of the forward process, which is imported as Lemma 8 and contributes the factor $k\log^2 T/T$. The DDIM coefficient $\eta^\star_t$ is chosen so that the Jacobian deviation of $\phi^\star_t$ stays small enough for this comparison to close.

What would settle it

For $p_{\mathrm{data}}$ uniform on a $k$-dimensional sphere embedded in $\mathbb{R}^d$, evaluate numerically the sum $\sum_{t=2}^T \mathbb{E}\|\Sigma_{\alpha_t}(X_t)\|_F^2$ under the paper's schedule (2.2); Theorem 1 requires this sum to stay below $C_6 k\log^2 T/T$, so any observed growth in the ambient dimension $d$ or a $k$-dependence worse than the stated one would directly contradict Lemma 8 and the theorem.

Watch

Extended reading notes

Core claim

The paper's central claim is Theorem 1: for the probability flow ODE sampler with the DDIM coefficient $\eta^\star_t = 1-\alpha_t-\sqrt{(1-\alpha_t)(\alpha_t-\alpha_t)}$, the total variation distance between the forward marginal $p_{X_1}$ and the reverse marginal $p_{Y_1}$ obeys $\operatorname{TV}(p_{X_1},p_{Y_1}) \le c(k+\log d)\log^3 T/T + c(\varepsilon_{\mathrm{score}}+\varepsilon_{\mathrm{Jacobi}})\log T$, where $k$ is the intrinsic dimension of $\operatorname{supp}(p_{\mathrm{data}})$ defined through metric entropy and $\varepsilon_{\mathrm{score}},\varepsilon_{\mathrm{Jacobi}}$ are averaged score and Jacobian estimation errors. With perfect score estimation this is $O(k/T)$ up to logarithms, replacing the ambient dimension in the previous $O(d/T)$ rate. The guarantee is stated for the time-1 marginals rather than the sampler's final output $Y_0$, and the paper attributes the improvement to the coefficient choice plus the metric-entropy notion of intrinsic dimension under a bounded-support assumption.

Load-bearing premise

The result depends on an imported lemma, not proved here, that the accumulated conditional covariance of the forward process is $O(k\log^2 T/T)$ under only bounded support; the theorem also bounds the time-1 marginal $p_{Y_1}$ rather than the sampler's final output $Y_0$, so either gap would leave the $O(k/T)$ guarantee short of the actual sampler.

Editorial extensions

If this is right

  • With accurate score estimation the probability flow ODE sampler converges at $O(k/T)$ in total variation, so it automatically exploits unknown low-dimensional structure without being told $k$.
  • This improves the previous $O(d/T)$ guarantee for this sampler and removes the requirement $T\gtrsim d^2$ that appeared in prior analysis.
  • Score error enters only through averaged $\ell^2$ score error and Jacobian error, each multiplied by $\log T$; no Hessian-level score control is required, in contrast to the concurrent bound.
  • The theorem controls $\operatorname{TV}(p_{X_1},p_{Y_1})$, the marginal one step before the final denoise, so it does not yet cover the actual output distribution of the full sampler.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Beyond the paper: if the final step from $Y_1$ to $Y_0$ is handled by one additional score-based update, the adaptive $O(k/T)$ guarantee would plausibly extend to generated samples; the paper leaves that extension implicit.
  • Beyond the paper: a testable prediction is that on synthetic data supported on a known low-dimensional manifold, a perfectly learned score yields reverse-marginal TV that decays with $T$ at the $k/T$ rate independent of the embedding dimension.
  • Beyond the paper: because the proof relies on an imported covariance lemma, non-smooth low-dimensional sets with large conditional covariance may violate the rate; checking the lemma on fractal or self-similar supports would clarify the true scope of adaptivity.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 4 minor

Summary. The paper studies the probability flow ODE sampler (a DDIM-type deterministic sampler) for score-based diffusion models and claims that, with the coefficient choice η*_t from the original DDIM paper, the sampler adapts to unknown low-dimensional structure of the target distribution. Theorem 1 states a total-variation bound of O((k + log d) log^3 T / T + (εscore + εJacobi) log T) between the forward marginal pX1 and the reverse marginal pY1, where k is the intrinsic dimension defined via metric entropy. The proof uses a high-probability set construction, a density-ratio recursion, and a lemma (Lemma 8) imported from a companion paper that supplies the sole k-dependent estimate. The paper also compares its result with a concurrent work by Liang et al. (2025), claiming that its own analysis does not require Hessian error control.

Significance. If fully established, the result would be a meaningful advance: it would show that a deterministic diffusion sampler automatically exploits low-dimensional structure, improving the ambient-dimension rate O(d/T) to O(k/T) (up to logarithms), and it would do so with a proof framework that is more concise than previous probability-flow analyses. The paper is clearly motivated, and the high-level proof strategy is interesting. However, two load-bearing issues prevent me from endorsing the paper in its current form: the theorem as stated does not control the sampler's actual output distribution, and the key dimension-dependent estimate is not proven in the manuscript.

major comments (2)
  1. [Section 3 (Theorem 1) and Section 1 (Abstract/Introduction)] Theorem 1 bounds TV(pX1, pY1), but the sampler defined in (2.6) iterates from t = T down to t = 1, so the generated sample is Y0, not Y1. The abstract states that "the probability flow ODE sampler achieves a convergence rate of O(k/T) in total variation distance," and the Introduction describes "the TV distance between the target and generated distributions." The proof in Section 4 analyzes only the density ratio at times t ≥ 1 and never studies the final deterministic map from Y1 to Y0. As written, the theorem does not support the advertised sampler-convergence claim; either an additional argument controlling TV(pX0, pY0) must be supplied, or the abstract and discussion must be revised to state the guarantee only for the intermediate marginal pY1.
  2. [Appendix B (Lemma 8) and Section 4.3 (step (b))] The O(k/T) rate is entirely carried by Lemma 8, which bounds ∑_t E ||√((1−α_t)/(α_t−\bar α_t)) ∂φ*_t(X_t)/∂x − I||_F² by C6 k log²T/T. The proof of Lemma 8 ends with "step (d) follows from (Li and Yan, 2025, Lemma 18)" without reproducing that lemma or its assumptions. Since the intrinsic dimension k enters the final rate only through this estimate, this is the central technical step of the paper. The bound is not a routine consequence of the preceding identities: by (A.8) the matrix equals (√(1+γ_t)−1)Σ_{α_t}(X_t), and its scaling with the schedule (2.2) is delicate. The authors should either prove Lemma 8 in full or state the imported lemma explicitly with all conditions and a proof, so that the reader can verify the k-dependence. Without this, the main theorem is not self-contained.
minor comments (4)
  1. [Section 2 (Set-up)] The notation α_t is used inconsistently: (2.2) defines α_t = 1 − β_t, while the text also uses α_t for the product ∏_{i=1}^t α_i (which should be \bar α_t). This creates confusion in (2.7), (4.1), (A.8), and Lemma 8, where expressions like "α_t − α_t" appear nonsensical as written.
  2. [Equation (2.7)] The displayed definition of η*_t appears to contain a typo: "1 − α_t − √((1−α_t)(α_t − α_t))" evaluates to a nonsensical expression; it should be written in terms of \bar α_t and either \bar α_{t−1} or the stepwise noise coefficient.
  3. [Section 1 (Introduction)] The phrase "T /greaterorsimilard2" should be the mathematical symbol "T ≳ d²" or similar.
  4. [Definition 1] The intrinsic dimension k is defined relative to a specific ε = T^{−cε}; this makes the notion of dimension T-dependent. This is not necessarily wrong, but it should be commented on, since the usual metric-entropy dimension would use k = O(log N_ε / log(1/ε)).

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the ODE sampler error is reduced to an imported forward-process covariance lemma, not to the theorem's own conclusion.

full rationale

Theorem 1 is not circular. The proof in Section 4 decomposes TV(pX1, pY1) into TV(pXT, pYT), sums of E[xi_t(X_t)], and probabilities of exiting the high-probability sets T_t, and the ODE-specific work is in Lemma 3 and Section 4.3, where the deterministic map phi_t is related to the exact and estimated score functions. The k-dependence in the final rate enters only through Lemma 8, which bounds the accumulated conditional covariance of the forward process and is imported without proof from (Li and Yan, 2025, Lemma 18), a self-citation by coauthor Y. Yan. This is a genuine verification gap and a correctness risk: if that cited lemma does not supply the stated k-bound under the stated schedule and Assumption 1, the adaptivity claim collapses. It is not, however, circular in the sense this review targets, because the cited lemma concerns the forward process's posterior covariance, not the probability-flow ODE's TV error; the paper does not assume its own conclusion. The appended skeptic concern that Lemma 8 may be scale-inconsistent for X0 ~ Uniform[0,1] would, if correct, make the lemma false under the stated assumptions, but that is a correctness objection rather than a reduction-by-definition. Minor self-citations, such as Lemma 1 from (Li and Yan, 2024), are technical imports and are not the source of the central adaptivity claim. The derivation chain is modular, not circular.

Assumptions & free parameters 0 free parameters · 6 assumptions · 0 invented entities

The paper introduces no fitted parameters: all constants (c0, c1, C1, C5, C6, Ccover, cR, cε) are universal and chosen sufficiently large, and the intrinsic dimension k is a property of the target distribution, not a fitted value. The central claim rests on several domain assumptions: bounded support, the covering-number definition of intrinsic dimension, the specific score-error metrics, and the cited technical lemmas, especially the unproved Lemma 8 from the authors' companion paper.

assumptions (6)
  • domain assumption Assumption 1: the support of pdata has radius bounded by T^{cR} for a universal constant cR
    Used throughout Lemma 2 and the proof of Lemma 8 to control the density ratio pYT/pXT; stated in Section 3 before Theorem 1.
  • domain assumption Intrinsic dimension k is defined via the covering-number bound log N_ε(X) ≤ Ccover k log T with ε = T^{-cε}
    Definition 1 in Section 3; this is the notion of low-dimensional structure that drives the O(k/T) rate.
  • domain assumption Score estimates satisfy the error measures in (2.4)-(2.5), including Jacobian error
    The theorem's error terms εscore and εJacobi are defined with respect to these metrics; the analysis assumes the learned score and its Jacobian are close in these averaged senses.
  • domain assumption The noise schedule (2.2) satisfies Lemma 5 (αT ≤ T^{-c2}, αt ≥ 1 - c1 log T / T, etc.)
    Lemma 5 is cited from Li et al. (2024b, Appendix A.2) and is used repeatedly, including in Lemma 2 and the proof of Lemma 3.
  • domain assumption Lemma 8: the accumulated conditional covariance of the forward process is bounded by O(k log^2 T / T)
    Stated as Lemma 8 in Appendix B, but the proof's final step cites (Li and Yan 2025, Lemma 18) without reproducing the argument; this is the key mechanism that makes the final rate scale with k.
  • standard math Standard Gaussian concentration and covering-net facts used to construct the high-probability sets Tt and G
    Used in Lemma 1 and in the proof of Lemma 3; these are textbook results (Vershynin 2018).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Adaptivity and Convergence of Probability Flow ODEs in Diffusion Generative Models." pith.science (2026). https://pith.science/paper/LRR7GI6N

@misc{pith2026250118863,
  author       = {Pith},
  title        = {Pith review of: Adaptivity and Convergence of Probability Flow ODEs in Diffusion Generative Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LRR7GI6N}},
  note         = {Machine review of arXiv:2501.18863}
}
abstract

Score-based generative models, which transform noise into data by learning to reverse a diffusion process, have become a cornerstone of modern generative AI. This paper contributes to establishing theoretical guarantees for the probability flow ODE, a widely used diffusion-based sampler known for its practical efficiency. While a number of prior works address its general convergence theory, it remains unclear whether the probability flow ODE sampler can adapt to the low-dimensional structures commonly present in natural image data. We demonstrate that, with accurate score function estimation, the probability flow ODE sampler achieves a convergence rate of $O(k/T)$ in total variation distance (ignoring logarithmic factors), where $k$ is the intrinsic dimension of the target distribution and $T$ is the number of iterations. This dimension-free convergence rate improves upon existing results that scale with the typically much larger ambient dimension, highlighting the ability of the probability flow ODE sampler to exploit intrinsic low-dimensional structures in the target distribution for faster sampling.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

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

  1. Faster Diffusion Models via Higher-Order Approximation

    cs.LG 2025-06 conditional novelty 7.0 of 10

    A new higher-order ODE sampler for diffusion models is proven to reach ε total-variation accuracy with eO(d^{1+2/K}/ε^{1/K}) iterations under mild assumptions.

  2. Fast Convergence for High-Order ODE Solvers in Diffusion Probabilistic Models

    cs.LG 2025-06 conditional novelty 6.0 of 10

    A TV convergence bound O(d^{7/4} ε^{1/2} + d(dH)^p) is proved for p-th order (exponential) Runge-Kutta samplers of probability-flow ODEs under C² smoothness of the learned score.

Reference graph

Works this paper leans on

45 extracted references · 12 canonical work pages · cited by 2 Pith papers

  1. [1]

    Anderson, B. D. (1982). Reverse-time diffusion equation models. Stochastic Processes and their Applications , 12(3):313--326

    1982

  2. [2]

    Azangulov, I., Deligiannidis, G., and Rousseau, J. (2024). Convergence of diffusion models under the manifold hypothesis in high-dimensions. arXiv preprint arXiv:2409.18804

    arXiv 2024

  3. [3]

    Benton, J., De Bortoli, V., Doucet, A., and Deligiannidis, G. (2023a). Linear convergence bounds for diffusion models via stochastic localization. arXiv preprint arXiv:2308.03686

    arXiv 2023

  4. [4]

    Benton, J., Deligiannidis, G., and Doucet, A. (2023b). Error bounds for flow matching methods. arXiv preprint arXiv:2305.16860

    arXiv 2023

  5. [5]

    Chen, H., Lee, H., and Lu, J. (2023a). Improved analysis of score-based generative modeling: User-friendly bounds under minimal smoothness assumptions. In International Conference on Machine Learning , pages 4735--4763. PMLR

    2023

  6. [6]

    Chen, M., Huang, K., Zhao, T., and Wang, M. (2023b). Score approximation, estimation and distribution recovery of diffusion models on low-dimensional data. In International Conference on Machine Learning , pages 4672--4712. PMLR

    work page 2023

  7. [7]

    Chen, S., Chewi, S., Lee, H., Li, Y., Lu, J., and Salim, A. (2023c). The probability flow ode is provably fast. arXiv preprint arXiv:2305.11798

    arXiv 2023

  8. [8]

    Chen, S., Chewi, S., Li, J., Li, Y., Salim, A., and Zhang, A. R. (2022). Sampling is as easy as learning the score: theory for diffusion models with minimal data assumptions. arXiv preprint arXiv:2209.11215

    arXiv 2022

Show all 45 references
  1. [9]

    Chen, S., Daras, G., and Dimakis, A. G. (2023d). Restoration-degradation beyond linear diffusions: A non-asymptotic analysis for DDIM -type samplers. arXiv preprint arXiv:2303.03384

    2023 arXiv

  2. [10]

    T., and Shah, M

    Croitoru, F.-A., Hondru, V., Ionescu, R. T., and Shah, M. (2023). Diffusion models in vision: A survey. IEEE Transactions on Pattern Analysis and Machine Intelligence

    2023

  3. [11]

    and Zhu, L

    Gao, X. and Zhu, L. (2024). Convergence analysis for general probability flow odes of diffusion models in wasserstein distances. arXiv preprint arXiv:2401.17958

    2024 arXiv

  4. [12]

    Haussmann, U. G. and Pardoux, E. (1986). Time reversal of diffusions. The Annals of Probability , pages 1188--1205

    1986

  5. [13]

    Ho, J., Jain, A., and Abbeel, P. (2020). Denoising diffusion probabilistic models. Advances in Neural Information Processing Systems , 33:6840--6851

    2020

  6. [14]

    G., Vignac, C., and Welling, M

    Hoogeboom, E., Satorras, V. G., Vignac, C., and Welling, M. (2022). Equivariant diffusion for molecule generation in 3d. In International conference on machine learning , pages 8867--8887. PMLR

    2022

  7. [15]

    Z., Huang, J., and Lin, Z

    Huang, D. Z., Huang, J., and Lin, Z. (2024a). Convergence analysis of probability flow ode for score-based generative models. arXiv preprint arXiv:2404.09730

    2024 arXiv

  8. [16]

    Huang, Z., Wei, Y., and Chen, Y. (2024b). Denoising diffusion probabilistic models are optimally adaptive to unknown low dimensionality. arXiv preprint arXiv:2410.18784

    2024

  9. [17]

    Hyv \"a rinen, A. (2005). Estimation of non-normalized statistical models by score matching. Journal of Machine Learning Research , 6(4)

    2005

  10. [18]

    Hyv \"a rinen, A. (2007). Some extensions of score matching. Computational statistics & data analysis , 51(5):2499--2512

    2007

  11. [19]

    Kong, Z., Ping, W., Huang, J., Zhao, K., and Catanzaro, B. (2021). DiffWave : A versatile diffusion model for audio synthesis. In International Conference on Learning Representations

    2021

  12. [20]

    Li, G., Huang, Y., Efimov, T., Wei, Y., Chi, Y., and Chen, Y. (2024a). Accelerating convergence of score-based diffusion models, provably. arXiv preprint arXiv:2403.03852

    2024 arXiv

  13. [21]

    Li, G., Wei, Y., Chen, Y., and Chi, Y. (2023). Towards non-asymptotic convergence for diffusion-based generative models. In The Twelfth International Conference on Learning Representations

    2023

  14. [22]

    Li, G., Wei, Y., Chi, Y., and Chen, Y. (2024b). A sharp convergence theory for the probability flow odes of diffusion models. arXiv preprint arXiv:2408.02320

    2024 arXiv

  15. [23]

    and Yan, Y

    Li, G. and Yan, Y. (2024). Adapting to unknown low-dimensional structures in score-based diffusion models. In The Thirty-eighth Annual Conference on Neural Information Processing Systems

    2024

  16. [24]

    and Yan, Y

    Li, G. and Yan, Y. (2025). O(d/T) convergence theory for diffusion probabilistic models under minimal assumptions. arXiv preprint arXiv:2409.18959, accepted to International Conference on Learning Representations

    2025 arXiv

  17. [25]

    Li, R., Di, Q., and Gu, Q. (2024c). Unified convergence analysis for score-based diffusion models with deterministic samplers. arXiv preprint arXiv:2410.14237

    2024 arXiv

  18. [26]

    Liang, J., Huang, Z., and Chen, Y. (2025). Low-dimensional adaptation of diffusion models: Convergence in total variation. arXiv preprint arXiv:2501.12982

    2025 arXiv

  19. [27]

    Oko, K., Akiyama, S., and Suzuki, T. (2023). Diffusion models are minimax optimal distribution estimators. arXiv preprint arXiv:2303.01861

    2023 arXiv

  20. [28]

    Pope, P., Zhu, C., Abdelkader, A., Goldblum, M., and Goldstein, T. (2021). The intrinsic dimension of images and its impact on learning. In 9th International Conference on Learning Representations, ICLR 2021, Virtual Event, Austria, May 3-7, 2021 . OpenReview.net

    2021

  21. [29]

    Potaptchik, P., Azangulov, I., and Deligiannidis, G. (2024). Linear convergence of diffusion models under the manifold hypothesis. arXiv preprint arXiv:2410.09046

    2024 arXiv

  22. [30]

    Ramesh, A., Dhariwal, P., Nichol, A., Chu, C., and Chen, M. (2022). Hierarchical text-conditional image generation with CLIP latents. arXiv preprint arXiv:2204.06125

    2022 arXiv

  23. [31]

    Rombach, R., Blattmann, A., Lorenz, D., Esser, P., and Ommer, B. (2022). High-resolution image synthesis with latent diffusion models. In IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 10684--10695

    2022

  24. [32]

    L., Ghasemipour, K., Gontijo Lopes, R., Karagol Ayan, B., Salimans, T., et al

    Saharia, C., Chan, W., Saxena, S., Li, L., Whang, J., Denton, E. L., Ghasemipour, K., Gontijo Lopes, R., Karagol Ayan, B., Salimans, T., et al. (2022). Photorealistic text-to-image diffusion models with deep language understanding. Advances in Neural Information Processing Sys...

    2022

  25. [33]

    Simoncelli, E. P. and Olshausen, B. A. (2001). Natural image statistics and neural representation. Annual review of neuroscience , 24(1):1193--1216

    2001

  26. [34]

    Song, J., Meng, C., and Ermon, S. (2020). Denoising diffusion implicit models. arXiv preprint arXiv:2010.02502

    2020 arXiv

  27. [35]

    and Ermon, S

    Song, Y. and Ermon, S. (2019). Generative modeling by estimating gradients of the data distribution. Advances in neural information processing systems , 32

    2019

  28. [36]

    P., Kumar, A., Ermon, S., and Poole, B

    Song, Y., Sohl-Dickstein, J., Kingma, D. P., Kumar, A., Ermon, S., and Poole, B. (2021). Score-based generative modeling through stochastic differential equations. International Conference on Learning Representations

    2021

  29. [37]

    Tang, R., Lin, L., and Yang, Y. (2024). Conditional diffusion models are minimax-optimal and manifold-adaptive for conditional distribution estimation. arXiv preprint arXiv:2409.20124

    2024 arXiv

  30. [38]

    and Yang, Y

    Tang, R. and Yang, Y. (2024). Adaptivity of diffusion models to manifold structures. In International Conference on Artificial Intelligence and Statistics , pages 1648--1656. PMLR

    2024

  31. [39]

    and Zhao, H

    Tang, W. and Zhao, H. (2024). Score-based diffusion models via stochastic differential equations--a technical tutorial. arXiv preprint arXiv:2402.07487

    2024 arXiv

  32. [40]

    Vershynin, R. (2018). High-dimensional probability: An introduction with applications in data science , volume 47. Cambridge university press

    2018

  33. [41]

    T., Castro, S., Kunze, J., and Erhan, D

    Villegas, R., Babaeizadeh, M., Kindermans, P.-J., Moraldo, H., Zhang, H., Saffar, M. T., Castro, S., Kunze, J., and Erhan, D. (2022). Phenaki: Variable length video generation from open domain textual descriptions. In International Conference on Learning Representations

    2022

  34. [42]

    Vincent, P. (2011). A connection between score matching and denoising autoencoders. Neural computation , 23(7):1661--1674

    2011

  35. [43]

    Wainwright, M. J. (2019). High-dimensional statistics: A non-asymptotic viewpoint , volume 48. Cambridge University Press

    2019

  36. [44]

    Wang, P., Zhang, H., Zhang, Z., Chen, S., Ma, Y., and Qu, Q. (2024). Diffusion models learn low-dimensional distributions via subspace clustering. arXiv preprint arXiv:2409.02426

    2024 arXiv

  37. [45]

    Yang, L., Zhang, Z., Song, Y., Hong, S., Xu, R., Zhao, Y., Shao, Y., Zhang, W., Cui, B., and Yang, M.-H. (2022). Diffusion models: A comprehensive survey of methods and applications. arXiv preprint arXiv:2209.00796

    2022

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.