Pith. sign in

REVIEW 3 major objections 4 minor 62 references

Disentangling Total-Variance and Signal-to-Noise-Ratio Improves Diffusion Models

T0 review · 3 major / 4 minor · reviewed 2026-08-08 · deepseek-v4-flash

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

Pith's one-line read This paper argues that diffusion-model sampling can be accelerated substantially by separating total variance from signal-to-noise ratio, that exploding-variance schedules are usually dominated by constant-variance variants with the same…

desk verdict A clean scheduling reformulation with a genuinely useful constant-TV insight, but the headline molecular gains rest on a model trained with a different schedule and need a retraining check. read the letter →

arxiv 2502.08598 v2 pith:BBF3RK6Z submitted 2025-02-12 cs.LG stat.ML

Khaled Kahouli , Winfried Ripken , Stefan Gugler , Oliver T. Unke , Klaus-Robert Müller , Shinichi Nakajima This is my paper · ORCID
classification cs.LGstat.ML
keywords diffusionmodelsnoisescheduletotalvariancesignal-to-noiseratioinversesigmoidSNRflowmatchingfew-stepsamplingmolecularstructuregeneration
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

This paper argues that the noise schedule of a diffusion model should be designed with two independent dials: the total variance (TV) of the perturbed signal and its signal-to-noise ratio (SNR). Working in that disentangled framework, the authors show that common schedules with exponentially growing total variance, such as SMLD and the uniform-time variant of EDM, are improved by switching to a constant TV while keeping the same SNR schedule. They then propose VP-ISSNR, a constant-TV schedule whose SNR decays like the exponential of an inverse sigmoid function, which generalizes optimal-transport flow matching. On QM9 molecular structure generation this schedule yields stable molecules with as few as four solver steps and state-of-the-art stability at 32 to 64 steps, with similar or better few-step image quality on CIFAR-10 and FFHQ. If the claim holds, sampling cost in diffusion models can be cut sharply without retraining, and schedule design becomes an axis that composes with better solvers.

What carries the argument

The central object is the TV/SNR parameterization of the perturbation kernel, which rewrites the forward noising distribution in terms of TV $\tau^2=a^2+b^2$ and SNR $\gamma=a/b$, with the corresponding SDE drift and diffusion coefficients derived from these two functions. This separation allows a schedule's noise magnitude and its signal retention to be changed independently. The proposed VP-ISSNR schedule fixes $\tau^2=1$ and sets $\gamma^2(t)=((1-t)/t)^{2\eta}e^{2\kappa}$, the exponential of an inverse sigmoid in $\log(1/t-1)$, which produces rapid SNR decay near both ends of the diffusion interval and reduces to the OTFM SNR when $\eta=1$, $\kappa=0$. The second load-bearing piece is the SNR-conditioned score network: the model is trained with constant TV and conditioned on a scaled, log-linearized SNR, $c_{\text{snr}}(\gamma^2)=\omega\log\gamma^2+\xi$, which lets one trained network be sampled under many schedules and makes the paper's cross-schedule comparisons possible.

What would settle it

Train a fresh score network from scratch under the VP-ISSNR schedule itself, with matched architecture and compute, and compare stability and FID at the same NFEs against the paper's numbers obtained by reusing a cosine-trained model. The appendix already hints that the cosine-trained model beats an EDM-trained model even when the EDM schedule is used for sampling; if consistent retraining removes or reverses the VP-ISSNR advantage, the schedule-specific claims would collapse. A cleaner test is to train and evaluate every candidate schedule under both matched and cross-schedule conditions and check whether the rank order of schedules survives matched training.

Watch

Extended reading notes

Core claim

The paper's central claim is that the total variance (TV), $\tau^2(t) = a^2(t)+b^2(t)$, and the signal-to-noise ratio (SNR), $\gamma(t)=a(t)/b(t)$, of a diffusion perturbation kernel are separate design choices, and that treating them as such improves fast sampling. Reformulating the kernel as $p(x(t)|x(0))=\mathcal{N}\big(\sqrt{\tau^2\gamma^2/(1+\gamma^2)}\;x(0),\; \tau^2/(1+\gamma^2)\;I\big)$, the authors show that schedules whose TV explodes exponentially, such as SMLD and EDM with a uniform time grid, are dominated at equal SNR by their constant-TV (VP) counterparts in both molecular and image generation. They propose the VP-ISSNR schedule, $\tau^2(t)=1$ with $\gamma^2(t)=((1-t)/t)^{2\eta}e^{2\kappa}$, an inverse-sigmoid SNR law that contains optimal-transport flow matching as a special case. With this schedule, a model trained once under a cosine SNR schedule generates stable QM9 molecules in four Euler steps, reaches roughly 87 percent stability at 128 steps, and exceeds 95 percent with stochastic sampling at 64 steps; on images it matches or improves the highly optimized EDM sampler at low step counts when the two schedule parameters are scaled or tuned. The paper also offers a mechanistic explanation from a toy mixture-of-delta experiment: good schedules keep ODE trajectories straight near the data manifold and grow the marginal support early.

Load-bearing premise

The evaluation assumes that a score network trained under one SNR schedule (a cosine schedule for molecules, pretrained EDM models for images) can be sampled under a different SNR schedule without retraining, so that the observed differences in stability and FID are caused by the schedule rather than by the training run.

Editorial extensions

If this is right

  • Schedules with exponentially exploding total variance (SMLD and uniform-time EDM) can be converted to constant-TV variants with the same SNR curve, improving or matching stability and FID at every tested number of function evaluations.
  • The VP-ISSNR schedule produces stable QM9 molecules with a first-order Euler ODE in just four NFEs, about 74 percent stability at eight NFEs, and nearly 87 percent at 128 NFEs; with the reverse SDE it reaches 93.16 percent at 32 NFEs and 95.82 percent at 64 NFEs, which the authors report as state of the art for those budgets.
  • The schedule benefit persists across Euler, Heun, DPM, and RK45 solvers, so noise-schedule optimization is orthogonal to solver improvements and can be combined with them.
  • On image datasets (CIFAR-10, FFHQ, AFHQ, ImageNet), VP-ISSNR with an NFE-scaled or tuned schedule parameter matches or improves EDM and OTFM FID in the low-NFE regime while reusing the same pretrained EDM score networks.
  • The toy trajectory analysis suggests a concrete design criterion for fast schedules: low ODE curvature near the data manifold at $t\approx 0$ and marginal support that saturates before $t=1$.
  • Because the molecular model is trained once under a cosine SNR schedule and then sampled under many schedules, the reported gains do not require retraining per schedule, provided the SNR-conditioned score transfer holds.

Reading between the lines

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

  • If the SNR-conditioned transfer is general, the TV/SNR split implies that existing pretrained diffusion checkpoints could be re-scheduled at inference time, per dataset or per NFE, without retraining; the paper demonstrates this capability for its own experiments but does not develop it into a general recipe for arbitrary checkpoints.
  • The inverse-sigmoid SNR allocates many integration steps near both $t\approx 0$ and $t\approx 1$, not just near the data manifold; a testable extension is that the high-noise end matters for high-frequency detail, which could be checked by ablating the offset parameter $\kappa$ while holding the total number of steps fixed.
  • The fact that VP-OTFM matches OTFM on molecules but underperforms on images hints that the optimal TV schedule is task-dependent; searching over $\tau(t)$ jointly with $\gamma(t)$ using the curvature and support diagnostics is a natural next step that the paper leaves open.
  • The curvature and support hypothesis could be turned into a training-free schedule optimizer: parametrize $\tau$ and $\gamma$, and minimize integrated trajectory curvature subject to a support-growth constraint on a small calibration set; the paper supplies the diagnostics but not the optimization.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 4 minor

Summary. The paper proposes a total-variance/signal-to-noise-ratio (TV/SNR) disentangled framework for diffusion noise schedules, in which the perturbation kernel is reparameterized by a TV-controlling function τ(t) and an SNR-controlling function γ(t). The authors derive forward/reverse SDEs for this parameterization, introduce constant-TV (VP) variants of existing VE schedules, and propose a new VP-ISSNR schedule based on the exponential of an inverse sigmoid SNR. They report improved few-step molecular structure generation on QM9 (stability rate as a function of NFE) and improved/competitive FID on CIFAR-10 and FFHQ, and supplement this with a numerical analysis of ODE trajectory curvature and marginal-support growth. The central claims are that exploding-TV schedules can be improved by constant-TV counterparts and that the proposed ISSNR schedule enables state-of-the-art few-step generation, especially for molecules.

Significance. If the claims hold, the TV/SNR reformulation is a useful unifying perspective, and the empirical gains in few-step molecular generation are practically significant. The manuscript ships an open-source implementation, recovers several existing schedules in a single table (Table A1), and includes a toy experiment on trajectory curvature and marginal support that goes beyond pure benchmarking. The VP-vs-VE comparisons on images and molecules use shared pretrained models, which is a reasonable first-order protocol, and Figure A4 provides a partial check of schedule transfer for molecules. However, the correctness of the derived SDE drift is compromised by an algebraic error, and the tuned image hyperparameters are selected per NFE without a stated validation protocol. These issues are load-bearing for the empirical conclusions, so the paper requires major revision before the central claims can be accepted.

major comments (3)
  1. [Appendix B.3, Eq. (42)] The drift coefficient is derived incorrectly. For a^2 = τ^2γ^2/(1+γ^2), differentiating log a gives f(t) = d/dt log a = τ'(t)/τ(t) + γ'(t)/[γ(t)(1+γ^2(t))], not τ'(t)/τ(t) + γ'(t)/γ(t) * (1+γ^2(t)). The last algebraic step in the derivation inverts the factor (1+γ^2). Consequently, the forward SDE and the reverse ODE/SDE used in Sections 3 and 4 are not the SDEs that correspond to the perturbation kernel in Eq. (8). For the pure VE case τ=1, γ=1/σ, the correct drift is f=0, whereas the paper's formula gives f = -σ'/σ (1+1/σ^2), a spurious non-zero drift. Since all sampling experiments, including the VP-ISSNR results, use this ODE/SDE, the authors must correct the derivation and re-run the experiments with the correct drift, or demonstrate that the code implemented the correct drift despite the erroneous derivation.
  2. [Section 4.2 and Figure 4] The 'ISSNR[tuned]' results are obtained by Bayesian optimization of η and κ separately for each NFE (32 trials), but the manuscript does not specify the selection criterion or whether the FID used for selection is computed on the same test set used for the reported numbers. Tuning directly against the reported FID would make the low-NFE improvements optimistic and not directly comparable with fixed-schedule baselines such as EDM and OTFM, whose hyperparameters were set in prior work. Please state the selection protocol (e.g., a validation split), and present the fixed 'scaled' variant as the primary image claim, with the tuned variant clearly labeled as an upper-bound/exploratory result.
  3. [Section 4.1 and Appendix C] The headline molecular results are produced by a single model trained with the DDPM-cos (ν=1) schedule and then evaluated with other schedules, relying on the network being conditioned on SNR rather than time. Figure A4 shows that a model trained with the EDM SNR schedule also benefits from VP-ISSNR sampling relative to other schedules, which is reassuring partial evidence. However, the manuscript never trains with the proposed VP-ISSNR schedule itself. Since schedule transfer is only as good as the score approximation over the queried (x̂, csnr(γ^2)) pairs, a decisive experiment that trains with VP-ISSNR (or demonstrates schedule-ranking invariance across more than two training schedules) is needed to support the claim that VP-ISSNR is a state-of-the-art scheduling choice for few-step molecular generation.
minor comments (4)
  1. [Section 3.3, Eq. (10)] The displayed formula for γ^2(t) is inconsistent with Table A1 and with the boundary expressions in Eqs. (11) and (12): as written, the argument of the logarithm becomes negative for t close to tmax (e.g., with tmax=0.99). The intended schedule appears to be γ^2(t) = ((1-t)/t)^{2η} exp(2κ) with a truncation/renormalization to the interval [tmin, tmax]; please repair the formula so that a reader can implement it directly.
  2. [Section 5, Conclusion] The statement of an 'up to 30-fold' improvement in molecular generation is ambiguous; the authors should specify the comparison point, e.g., the NFE at which the best baseline reaches the same stability as VP-ISSNR at 4 or 8 NFEs.
  3. [Section 4.3] The curvature measure E[||(x(1)-x(0))-ẋ(t)||^2] is used to support the trajectory-straightness hypothesis, but the manuscript does not explain why this quantity controls the discretization error of the reverse ODE; a short derivation or a reference would make the argument easier to evaluate.
  4. [Appendix A, Eq. (7)] The definition of the EDM non-uniform time grid uses σ_{i<N} and then sets σ_N = 0, which is standard but can be confusing; adding an explicit note that σ_N = 0 corresponds to the data endpoint would improve readability.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: the TV/SNR reparameterization is algebraic and the schedule improvements are empirical external-benchmark results; the reuse of a single trained network for schedule comparisons is a limitation but not a definitional reduction.

full rationale

The paper's core formalism, Eq. (8), is an exact algebraic reparameterization of the perturbation kernel in Eq. (1): comparing coefficients gives TV = a^2 + b^2 = tau^2 and SNR = a/b = gamma, so no performance claim is derived from the definition itself. The proposed VP variants are new schedules (tau=1 with unchanged gamma), and their reported superiority is an empirical result measured by QM9 stability rates and CIFAR-10/FFHQ FID scores against external baselines (EDM, OTFM, SMLD, DDPM variants). The VP-ISSNR schedule is explicitly defined in Eqs. (9)-(10), with fixed parameters eta=1.0, kappa=2.0 for molecules; the image-generation variants are transparently labeled as 'scaled' or 'tuned' (Bayesian optimization over 32 trials), so the tuned FID numbers are not disguised predictions. Self-citations to Kahouli et al. (2024) supply only the architecture and data split, not a uniqueness theorem or the schedule claim, so no load-bearing self-citation chain exists. The main weakness is experimental rather than circular: Appendix C states that all molecular models are trained with tau(t)=1 and DDPM-cos, and that this makes the model 'compatible with various TV and SNR schedules during sampling without retraining.' Thus the VP-vs-non-VP molecular comparisons use a single network trained under a constant-TV schedule, and Figure A4 retrains only with the EDM SNR schedule (still tau=1), never with VP-ISSNR itself. If score-network transfer across tau schedules is imperfect, the measured gains could partly reflect training-schedule alignment rather than the TV/SNR geometry. However, this is a train/sample mismatch confound, not a case where the conclusion equals its input by construction; the image experiments use pretrained EDM models trained under a non-constant-TV VE schedule and still show VP variants improving, providing independent evidence. The curvature-and-support analysis (Section 4.3) is explicitly presented as a post-hoc hypothesis, not a derivation of performance from the framework. Accordingly, no step satisfies the quoted-reduction test for circularity.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The central claim rests on two empirical premises: the validity of schedule transfer without retraining, and the choice of the ISSNR hyperparameters. No new physical or mathematical entities are introduced.

free parameters (5)
  • eta (steepness of ISSNR) = 1.0 for molecules; scaled with NFE or Bayesian-tuned per NFE for images
    Controls the steepness of the inverse-sigmoid SNR curve in Eq. (10) and directly shapes the schedule.
  • kappa (offset of ISSNR) = 2.0 for molecules; 0 or Bayesian-tuned for images
    Offsets the SNR schedule in Eq. (10); tuned values affect generation quality.
  • tmin = 0.01 for molecules; not stated for images
    Set to the largest value giving almost noiseless samples, per Appendix C; affects the gamma_max boundary.
  • tmax = 0.99 for molecules; tuned to approximate dataset maximum pairwise distance
    Sets the gamma_min boundary and is fitted to the dataset in Appendix C.
  • omega, xi (csnr linearization) = 0.35, -0.125
    Constants in the scaled-SNR conditioning used for schedule transfer during sampling (Appendix C).
assumptions (4)
  • domain assumption The denoising score-matching objective in Eq. (5) yields an accurate epsilon prediction for all schedules used at sampling.
    The paper trains one model per task and then evaluates many schedules with it, which requires the learned score to transfer across schedules.
  • domain assumption Data are standardized to zero mean and unit variance.
    Section 2.1 states this normalization and it underlies the TV/SNR parameterization and unit-variance input scaling.
  • ad hoc to paper A model trained with DDPM-cos can be evaluated with other schedules, including VP-ISSNR, without retraining.
    Appendix C explains that the model is trained on DDPM-cos and then sampled with all schedules; no VP-ISSNR-trained model is tested.
  • standard math Standard Ito calculus and Tweedie's formula are valid for deriving the SDE and analyzing trajectories.
    Appendix B uses the Ito integral solutions; Section 4.3 invokes Tweedie's formula.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Disentangling Total-Variance and Signal-to-Noise-Ratio Improves Diffusion Models." pith.science (2026). https://pith.science/paper/BBF3RK6Z

@misc{pith2026250208598,
  author       = {Pith},
  title        = {Pith review of: Disentangling Total-Variance and Signal-to-Noise-Ratio Improves Diffusion Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BBF3RK6Z}},
  note         = {Machine review of arXiv:2502.08598}
}
abstract

The long sampling time of diffusion models remains a significant bottleneck, which can be mitigated by reducing the number of diffusion time steps. However, the quality of samples with fewer steps is highly dependent on the noise schedule, i.e., the specific manner in which noise is introduced and the signal is reduced at each step. Although prior work has improved upon the original variance-preserving and variance-exploding schedules, these approaches $\textit{passively}$ adjust the total variance, without direct control over it. In this work, we propose a novel total-variance/signal-to-noise-ratio disentangled (TV/SNR) framework, where TV and SNR can be controlled independently. Our approach reveals that schedules where the TV explodes exponentially can often be improved by adopting a constant TV schedule while preserving the same SNR schedule. Furthermore, generalizing the SNR schedule of the optimal transport flow matching significantly improves the generation performance. Our findings hold across various reverse diffusion solvers and diverse applications, including molecular structure and image generation.

Figures

Figures reproduced from arXiv: 2502.08598 by the authors.

Figure 1
Figure 1. A: Our VP-ISSNR schedule enables the generation of stable molecules (top) with diffusion models using far fewer numbers of function evaluations (NFEs) than established methods like EDM and OTFM. It also converges to visually appealing images (bottom) faster than the optimized EDM sampler, and comparably to OTFM. B: A numerical analysis of ODE trajectories reveals that our proposed VP-ISSNR schedule has both low traj… view at source ↗
Figure 2
Figure 2. TV (left) and SNR (middle) schedules of different established scheduling methods (see Table A1 in Appendix A [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Stability rate (higher is better) as a function of the number of function evaluations (NFE) for molecular structure generation on the QM9 dataset. Left: Comparison between commonly established non-VP schedules (i.e., τ (t) non-constant) and their VP counterparts in our framework. Middle: Comparison of various baselines, including the VP analogs from the left plot, against our proposed VP-ISSNR schedule, with fixed p… view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: FID score (lower is better) as a function of the number of function evaluations (NFE) for image generation on CIFAR-10 and FFHQ. Left: Comparison of existing non-VP models and their VP variants on CIFAR-10. Middle: Comparison on CIFAR-10 between baseline methods and tw…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

62 extracted references · 37 canonical work pages

  1. [1]

    Deep unsupervised learning using nonequilibrium thermodynamics

    Jascha Sohl-Dickstein, Eric Weiss, Niru Maheswaranathan, and Surya Ganguli. Deep unsupervised learning using nonequilibrium thermodynamics. In Francis Bach and David Blei, editors, Proceedings of the 32nd International Conference on Machine Learning, volume 37 of Proceedings of Machine Learning Research, pages 2256--2265, Lille, France, 07--09 Jul 2015. P...

    2015

  2. [2]

    Denoising diffusion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffusion probabilistic models. In H. Larochelle, M. Ranzato, R. Hadsell, M.F. Balcan, and H. Lin, editors, Advances in Neural Information Processing Systems, volume 33, pages 6840--6851. Curran Associates, Inc., 2020. URL https://proceedings.neurips.cc/paper_files/paper/2020/file/4c5bcfec8584af0d967f1a...

    2020

  3. [3]

    Improved techniques for training score-based generative models

    Yang Song and Stefano Ermon. Improved techniques for training score-based generative models. In H. Larochelle, M. Ranzato, R. Hadsell, M.F. Balcan, and H. Lin, editors, Advances in Neural Information Processing Systems, volume 33, pages 12438--12448. Curran Associates, Inc., 2020. URL https://proceedings.neurips.cc/paper_files/paper/2020/file/92c3b916311a...

    work page 2020

  4. [4]

    Score-based generative modeling through stochastic differential equations

    Yang Song, Jascha Sohl-Dickstein, Diederik P Kingma, Abhishek Kumar, Stefano Ermon, and Ben Poole. Score-based generative modeling through stochastic differential equations. In International Conference on Learning Representations, 2021 a . URL https://openreview.net/forum?id=PxTIG12RRHS

    2021

  5. [5]

    Diffusion models beat GAN s on image synthesis

    Prafulla Dhariwal and Alexander Quinn Nichol. Diffusion models beat GAN s on image synthesis. In A. Beygelzimer, Y. Dauphin, P. Liang, and J. Wortman Vaughan, editors, Advances in Neural Information Processing Systems, 2021. URL https://openreview.net/forum?id=AAWuCvzaVt

    2021

  6. [6]

    GLIDE : Towards photorealistic image generation and editing with text-guided diffusion models

    Alexander Quinn Nichol, Prafulla Dhariwal, Aditya Ramesh, Pranav Shyam, Pamela Mishkin, Bob Mcgrew, Ilya Sutskever, and Mark Chen. GLIDE : Towards photorealistic image generation and editing with text-guided diffusion models. In Kamalika Chaudhuri, Stefanie Jegelka, Le Song, Csaba Szepesvari, Gang Niu, and Sivan Sabato, editors, Proceedings of the 39th In...

    work page 2022

  7. [7]

    High-resolution image synthesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bj \"o rn Ommer. High-resolution image synthesis with latent diffusion models. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (CVPR), pages 10684--10695, 2022

    work page 2022

  8. [8]

    Scalable diffusion models with transformers

    William Peebles and Saining Xie. Scalable diffusion models with transformers. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pages 4195--4205, October 2023

    work page 2023

Show all 62 references
  1. [9]

    Albergo, Nicholas M

    Nanye Ma, Mark Goldstein, Michael S. Albergo, Nicholas M. Boffi, Eric Vanden-Eijnden, and Saining Xie. Sit: Exploring flow and diffusion-based generative models with scalable interpolant transformers. In Ale s Leonardis, Elisa Ricci, Stefan Roth, Olga Russakovsky, Torsten Satt...

    2024

  2. [10]

    Diffwave: A versatile diffusion model for audio synthesis

    Zhifeng Kong, Wei Ping, Jiaji Huang, Kexin Zhao, and Bryan Catanzaro. Diffwave: A versatile diffusion model for audio synthesis. In International Conference on Learning Representations, 2021. URL https://openreview.net/forum?id=a-xFK8Ymz5J

    2021

  3. [11]

    Wavegrad: Estimating gradients for waveform generation

    Nanxin Chen, Yu Zhang, Heiga Zen, Ron J Weiss, Mohammad Norouzi, and William Chan. Wavegrad: Estimating gradients for waveform generation. In International Conference on Learning Representations, 2021. URL https://openreview.net/forum?id=NsMLjcFaO8O

    2021

  4. [12]

    A udio LDM : Text-to-audio generation with latent diffusion models

    Haohe Liu, Zehua Chen, Yi Yuan, Xinhao Mei, Xubo Liu, Danilo Mandic, Wenwu Wang, and Mark D Plumbley. A udio LDM : Text-to-audio generation with latent diffusion models. In Andreas Krause, Emma Brunskill, Kyunghyun Cho, Barbara Engelhardt, Sivan Sabato, and Jonathan Scarlett, ...

    2023

  5. [13]

    u ller, and Kristof T Sch \

    Niklas WA Gebauer, Michael Gastegger, Stefaan SP Hessmann, Klaus-Robert M \"u ller, and Kristof T Sch \"u tt. Inverse design of 3d molecular structures with conditional generative neural networks. Nature communications, 13: 0 973, 2022

    2022

  6. [14]

    Equivariant diffusion for molecule generation in 3 D

    Emiel Hoogeboom, V\' ctor Garcia Satorras, Cl \'e ment Vignac, and Max Welling. Equivariant diffusion for molecule generation in 3 D . In Kamalika Chaudhuri, Stefanie Jegelka, Le Song, Csaba Szepesvari, Gang Niu, and Sivan Sabato, editors, Proceedings of the 39th International...

    2022

  7. [15]

    Diffusion-based molecule generation with informative prior bridges

    Lemeng Wu, Chengyue Gong, Xingchao Liu, Mao Ye, and Qiang Liu. Diffusion-based molecule generation with informative prior bridges. In S. Koyejo, S. Mohamed, A. Agarwal, D. Belgrave, K. Cho, and A. Oh, editors, Advances in Neural Information Processing Systems, volume 35, pages...

    2022

  8. [16]

    Mdm: Molecular diffusion model for 3d molecule generation

    Lei Huang, Hengtong Zhang, Tingyang Xu, and Ka-Chun Wong. Mdm: Molecular diffusion model for 3d molecule generation. Proceedings of the AAAI Conference on Artificial Intelligence, 37 0 (4): 0 5105--5112, Jun. 2023. doi:10.1609/aaai.v37i4.25639. URL https://ojs.aaai.org/index.p...

    2023 doi

  9. [17]

    Dror, Stefano Ermon, and Jure Leskovec

    Minkai Xu, Alexander S Powers, Ron O. Dror, Stefano Ermon, and Jure Leskovec. Geometric latent diffusion models for 3 D molecule generation. In Andreas Krause, Emma Brunskill, Kyunghyun Cho, Barbara Engelhardt, Sivan Sabato, and Jonathan Scarlett, editors, Proceedings of the 4...

  10. [18]

    M ol D iff: Addressing the atom-bond inconsistency problem in 3 D molecule diffusion generation

    Xingang Peng, Jiaqi Guan, Qiang Liu, and Jianzhu Ma. M ol D iff: Addressing the atom-bond inconsistency problem in 3 D molecule diffusion generation. In Andreas Krause, Emma Brunskill, Kyunghyun Cho, Barbara Engelhardt, Sivan Sabato, and Jonathan Scarlett, editors, Proceedings...

    2023

  11. [19]

    Midi: Mixed graph and 3d denoising diffusion for molecule generation

    Clement Vignac, Nagham Osman, Laura Toni, and Pascal Frossard. Midi: Mixed graph and 3d denoising diffusion for molecule generation. In ICLR 2023 - Machine Learning for Drug Discovery workshop, 2023. URL https://openreview.net/forum?id=M6Ifac3G4HK

    2023

  12. [20]

    Navigating the design space of equivariant diffusion-based generative models for de novo 3d molecule generation

    Tuan Le, Julian Cremer, Frank Noe, Djork-Arn \'e Clevert, and Kristof T Sch \"u tt. Navigating the design space of equivariant diffusion-based generative models for de novo 3d molecule generation. In The Twelfth International Conference on Learning Representations, 2024. URL h...

    2024

  13. [21]

    Molecular relaxation by reverse diffusion with time step prediction

    Khaled Kahouli, Stefaan Simon Pierre Hessmann, Klaus-Robert Müller, Shinichi Nakajima, Stefan Gugler, and Niklas Wolf Andreas Gebauer. Molecular relaxation by reverse diffusion with time step prediction. Machine Learning: Science and Technology, 5 0 (3): 0 035038, aug 2024. do...

    2024 doi

  14. [22]

    GeoDiff : A geometric diffusion model for molecular conformation generation

    Minkai Xu, Lantao Yu, Yang Song, Chence Shi, Stefano Ermon, and Jian Tang. GeoDiff : A geometric diffusion model for molecular conformation generation. In International Conference on Learning Representations, 2022. URL https://openreview.net/forum?id=PzcvxEMzvQC

    2022

  15. [23]

    Improved denoising diffusion probabilistic models

    Alexander Quinn Nichol and Prafulla Dhariwal. Improved denoising diffusion probabilistic models. In Marina Meila and Tong Zhang, editors, Proceedings of the 38th International Conference on Machine Learning, volume 139 of Proceedings of Machine Learning Research, pages 8162--8...

    2021

  16. [24]

    Learning fast samplers for diffusion models by differentiating through sample quality

    Daniel Watson, William Chan, Jonathan Ho, and Mohammad Norouzi. Learning fast samplers for diffusion models by differentiating through sample quality. In International Conference on Learning Representations, 2022. URL https://openreview.net/forum?id=VFBjuF8HEp

    2022

  17. [25]

    Progressive distillation for fast sampling of diffusion models

    Tim Salimans and Jonathan Ho. Progressive distillation for fast sampling of diffusion models. In International Conference on Learning Representations, 2022. URL https://openreview.net/forum?id=TIdIXIpzhoI

    2022

  18. [26]

    Consistency models

    Yang Song, Prafulla Dhariwal, Mark Chen, and Ilya Sutskever. Consistency models. In Andreas Krause, Emma Brunskill, Kyunghyun Cho, Barbara Engelhardt, Sivan Sabato, and Jonathan Scarlett, editors, Proceedings of the 40th International Conference on Machine Learning, volume 202...

    2023

  19. [27]

    Simplifying, stabilizing and scaling continuous-time consistency models

    Cheng Lu and Yang Song. Simplifying, stabilizing and scaling continuous-time consistency models. In The Thirteenth International Conference on Learning Representations, 2025. URL https://openreview.net/forum?id=LyJi5ugyJx

    2025

  20. [28]

    Simple reflow: Improved techniques for fast flow models

    Beomsu Kim, Yu-Guan Hsieh, Michal Klein, marco cuturi, Jong Chul Ye, Bahjat Kawar, and James Thornton. Simple reflow: Improved techniques for fast flow models. In The Thirteenth International Conference on Learning Representations, 2025. URL https://openreview.net/forum?id=fpvgSDKXGY

    2025

  21. [29]

    DPM -solver: A fast ODE solver for diffusion probabilistic model sampling in around 10 steps

    Cheng Lu, Yuhao Zhou, Fan Bao, Jianfei Chen, Chongxuan Li, and Jun Zhu. DPM -solver: A fast ODE solver for diffusion probabilistic model sampling in around 10 steps. In Alice H. Oh, Alekh Agarwal, Danielle Belgrave, and Kyunghyun Cho, editors, Advances in Neural Information Pr...

    2022

  22. [30]

    Pseudo numerical methods for diffusion models on manifolds

    Luping Liu, Yi Ren, Zhijie Lin, and Zhou Zhao. Pseudo numerical methods for diffusion models on manifolds. In International Conference on Learning Representations, 2022. URL https://openreview.net/forum?id=PlKWVd2yBkY

    2022

  23. [31]

    GENIE : Higher-order denoising diffusion solvers

    Tim Dockhorn, Arash Vahdat, and Karsten Kreis. GENIE : Higher-order denoising diffusion solvers. In Alice H. Oh, Alekh Agarwal, Danielle Belgrave, and Kyunghyun Cho, editors, Advances in Neural Information Processing Systems, 2022. URL https://openreview.net/forum?id=LKEYuYNOqx

    2022

  24. [32]

    Gotta go fast when generating data with score-based models, 2022

    Alexia Jolicoeur-Martineau, Ke Li, R \'e mi Pich \'e -Taillefer, Tal Kachman, and Ioannis Mitliagkas. Gotta go fast when generating data with score-based models, 2022. URL https://openreview.net/forum?id=YmONQIWli--

    2022

  25. [33]

    DPM -solver-v3: Improved diffusion ODE solver with empirical model statistics

    Kaiwen Zheng, Cheng Lu, Jianfei Chen, and Jun Zhu. DPM -solver-v3: Improved diffusion ODE solver with empirical model statistics. In Thirty-seventh Conference on Neural Information Processing Systems, 2023. URL https://openreview.net/forum?id=9fWKExmKa0

    2023

  26. [34]

    Fast sampling of diffusion models with exponential integrator

    Qinsheng Zhang and Yongxin Chen. Fast sampling of diffusion models with exponential integrator. In The Eleventh International Conference on Learning Representations, 2023. URL https://openreview.net/forum?id=Loek7hfb46P

    2023

  27. [35]

    Uni PC : A unified predictor-corrector framework for fast sampling of diffusion models

    Wenliang Zhao, Lujia Bai, Yongming Rao, Jie Zhou, and Jiwen Lu. Uni PC : A unified predictor-corrector framework for fast sampling of diffusion models. In Thirty-seventh Conference on Neural Information Processing Systems, 2023. URL https://openreview.net/forum?id=hrkmlPhp1u

    2023

  28. [36]

    On the importance of noise scheduling for diffusion models

    Ting Chen. On the importance of noise scheduling for diffusion models. arXiv preprint arXiv:2301.10972, 2023

    2023 arXiv

  29. [37]

    Common diffusion noise schedules and sample steps are flawed

    Shanchuan Lin, Bingchen Liu, Jiashi Li, and Xiao Yang. Common diffusion noise schedules and sample steps are flawed. In Proceedings of the IEEE/CVF winter conference on applications of computer vision, pages 5404--5411, 2024

    2024

  30. [38]

    Elucidating the design space of diffusion-based generative models

    Tero Karras, Miika Aittala, Timo Aila, and Samuli Laine. Elucidating the design space of diffusion-based generative models. In Alice H. Oh, Alekh Agarwal, Danielle Belgrave, and Kyunghyun Cho, editors, Advances in Neural Information Processing Systems, 2022. URL https://openre...

    2022

  31. [39]

    Flow straight and fast: Learning to generate and transfer data with rectified flow

    Xingchao Liu, Chengyue Gong, and qiang liu. Flow straight and fast: Learning to generate and transfer data with rectified flow. In The Eleventh International Conference on Learning Representations, 2023 b . URL https://openreview.net/forum?id=XVjTT1nw5z

    2023

  32. [40]

    Yaron Lipman, Ricky T. Q. Chen, Heli Ben-Hamu, Maximilian Nickel, and Matthew Le. Flow matching for generative modeling. In The Eleventh International Conference on Learning Representations, 2023. URL https://openreview.net/forum?id=PqvMRDCJT9t

    2023

  33. [41]

    Building normalizing flows with stochastic interpolants

    Michael Samuel Albergo and Eric Vanden-Eijnden. Building normalizing flows with stochastic interpolants. In The Eleventh International Conference on Learning Representations, 2023. URL https://openreview.net/forum?id=li7qeBbCR1t

    2023

  34. [42]

    Aram-Alexandre Pooladian, Heli Ben-Hamu, Carles Domingo-Enrich, Brandon Amos, Yaron Lipman, and Ricky T. Q. Chen. Multisample flow matching: Straightening flows with minibatch couplings. In Andreas Krause, Emma Brunskill, Kyunghyun Cho, Barbara Engelhardt, Sivan Sabato, and Jo...

    2023

  35. [43]

    Improving and generalizing flow-based generative models with minibatch optimal transport

    Alexander Tong, Kilian FATRAS, Nikolay Malkin, Guillaume Huguet, Yanlei Zhang, Jarrid Rector-Brooks, Guy Wolf, and Yoshua Bengio. Improving and generalizing flow-based generative models with minibatch optimal transport. Transactions on Machine Learning Research, 2024. ISSN 283...

    2024

  36. [44]

    Equivariant flow matching

    Leon Klein, Andreas Kr \"a mer, and Frank Noe. Equivariant flow matching. In Thirty-seventh Conference on Neural Information Processing Systems, 2023. URL https://openreview.net/forum?id=eLH2NFOO1B

    2023

  37. [45]

    Equivariant flow matching with hybrid probability transport for 3d molecule generation

    Yuxuan Song, Jingjing Gong, Minkai Xu, Ziyao Cao, Yanyan Lan, Stefano Ermon, Hao Zhou, and Wei-Ying Ma. Equivariant flow matching with hybrid probability transport for 3d molecule generation. In Thirty-seventh Conference on Neural Information Processing Systems, 2023 b . URL h...

    2023

  38. [46]

    Efficient 3d molecular generation with flow matching and scale optimal transport

    Ross Irwin, Alessandro Tibo, Jon Paul Janet, and Simon Olsson. Efficient 3d molecular generation with flow matching and scale optimal transport. In ICML 2024 AI for Science Workshop, 2024. URL https://openreview.net/forum?id=CxAjGjdkqu

    2024

  39. [47]

    Generative modeling by estimating gradients of the data distribution

    Yang Song and Stefano Ermon. Generative modeling by estimating gradients of the data distribution. In H. Wallach, H. Larochelle, A. Beygelzimer, F. d Alch\' e -Buc, E. Fox, and R. Garnett, editors, Advances in Neural Information Processing Systems, volume 32. Curran Associates...

    2019

  40. [48]

    Variational diffusion models

    Diederik P Kingma, Tim Salimans, Ben Poole, and Jonathan Ho. Variational diffusion models. In A. Beygelzimer, Y. Dauphin, P. Liang, and J. Wortman Vaughan, editors, Advances in Neural Information Processing Systems, 2021. URL https://openreview.net/forum?id=2LdBqxc1Yv

    2021

  41. [49]

    Diffusion normalizing flow

    Qinsheng Zhang and Yongxin Chen. Diffusion normalizing flow. In A. Beygelzimer, Y. Dauphin, P. Liang, and J. Wortman Vaughan, editors, Advances in Neural Information Processing Systems, 2021. URL https://openreview.net/forum?id=x1Lp2bOlVIo

    2021

  42. [50]

    A connection between score matching and denoising autoencoders

    Pascal Vincent. A connection between score matching and denoising autoencoders. Neural Computation, 23 0 (7): 0 1661--1674, 2011. doi:10.1162/NECO_a_00142

    2011 doi

  43. [51]

    Denoising diffusion implicit models

    Jiaming Song, Chenlin Meng, and Stefano Ermon. Denoising diffusion implicit models. In International Conference on Learning Representations, 2021 b . URL https://openreview.net/forum?id=St1giarCHLP

    2021

  44. [52]

    DPM -solver: A fast ODE solver for diffusion probabilistic model sampling in around 10 steps

    Cheng Lu, Yuhao Zhou, Fan Bao, Jianfei Chen, Chongxuan Li, and Jun Zhu. DPM -solver: A fast ODE solver for diffusion probabilistic model sampling in around 10 steps. Advances in Neural Information Processing Systems, 35: 0 5775--5787, 2022 b

    2022

  45. [53]

    Equivariant flow matching for molecular conformer generation

    Majdi Hassan, Nikhil Shenoy, Jungyoon Lee, Hannes Stark, Stephan Thaler, and Dominique Beaini. Equivariant flow matching for molecular conformer generation. In ICML'24 Workshop ML for Life and Material Science: From Theory to Industry Applications, 2024. URL https://openreview...

    2024

  46. [54]

    Ramakrishnan, P

    R. Ramakrishnan, P. O. Dral, M. Rupp, and O. A. von Lilienfeld. Quantum chemistry structures and properties of 134 kilo molecules. Scientific Data, 1 0 (1): 0 140022, 2014. doi:10.1038/sdata.2014.22

    2014 doi

  47. [55]

    DPM - Solver ++: Fast Solver for Guided Sampling of Diffusion Probabilistic Models , May 2023

    Cheng Lu, Yuhao Zhou, Fan Bao, Jianfei Chen, Chongxuan Li, and Jun Zhu. DPM - Solver ++: Fast Solver for Guided Sampling of Diffusion Probabilistic Models , May 2023. URL http://arxiv.org/abs/2211.01095. arXiv:2211.01095 [cs]

    2023 arXiv

  48. [56]

    Gans trained by a two time-scale update rule converge to a local nash equilibrium

    Martin Heusel, Hubert Ramsauer, Thomas Unterthiner, Bernhard Nessler, and Sepp Hochreiter. Gans trained by a two time-scale update rule converge to a local nash equilibrium. In I. Guyon, U. Von Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, and R. Garnett, editors...

    2017

  49. [57]

    Learning multiple layers of features from tiny images

    Alex Krizhevsky. Learning multiple layers of features from tiny images. Technical Report 0, University of Toronto, Toronto, Ontario, 2009. URL https://www.cs.toronto.edu/ kriz/learning-features-2009-TR.pdf

    2009

  50. [58]

    A style-based generator architecture for generative adversarial networks

    Tero Karras, Samuli Laine, and Timo Aila. A style-based generator architecture for generative adversarial networks. 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 4396--4405, 2018. URL https://openaccess.thecvf.com/content_CVPR_2019/papers/Ka...

    2019

  51. [59]

    Stargan v2: Diverse image synthesis for multiple domains

    Yunjey Choi, Youngjung Uh, Jaejun Yoo, and Jung-Woo Ha. Stargan v2: Diverse image synthesis for multiple domains. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 2020

    2020

  52. [60]

    Imagenet: A large-scale hierarchical image database

    Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li, and Li Fei-Fei. Imagenet: A large-scale hierarchical image database. In 2009 IEEE Conference on Computer Vision and Pattern Recognition, pages 248--255, 2009. doi:10.1109/CVPR.2009.5206848

    2009

  53. [61]

    Tweedie’s formula and selection bias

    Bradley Efron. Tweedie’s formula and selection bias. Journal of the American Statistical Association, 106 0 (496): 0 1602--1614, 2011. doi:10.1198/jasa.2011.tm11181. URL https://doi.org/10.1198/jasa.2011.tm11181. PMID: 22505788

    2011 doi

  54. [62]

    Applied Stochastic Differential Equations

    Simo Särkkä and Arno Solin. Applied Stochastic Differential Equations. Cambridge University Press, 1 edition, April 2019. ISBN 978-1-108-18673-5. doi:10.1017/9781108186735. URL https://www.cambridge.org/core/product/identifier/9781108186735/type/book

    2019

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.