Pith. sign in

REVIEW 4 major objections 6 minor 1 cited by

Physics-Informed Distillation of Diffusion Models for PDE-Constrained Generation

T0 review · 4 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read Post-hoc distillation enforces PDE laws on final diffusion samples, avoiding the Jensen's Gap and enabling one-step generation.

desk verdict Useful distillation trick for physics-constrained diffusion, but the headline PDE-error gain is partly circular; worth a rigorous referee. read the letter →

arxiv 2505.22391 v2 pith:FZNLCDIC submitted 2025-05-28 cs.LG cs.AIcs.CEcs.NAmath.NA

Yi Zhang , Peng Wang , Difan Zou This is my paper
classification cs.LGcs.AIcs.CEcs.NAmath.NA
keywords diffusionmodelsphysics-informeddistillationPDE-constrainedgenerationJensen'sgapone-stepforwardandinverseproblemspartialobservationreconstructionfinite-differenceresidual
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

The paper claims that the standard way of injecting PDE constraints into diffusion models—evaluating the residual on the denoising estimate $\mathbb{E}[x_0|x_t]$—is theoretically mismatched, because a nonlinear operator applied to an average is not the average of the operator applied to samples. It proposes to sidestep this Jensen's Gap by distilling the diffusion model into a one-step student that maps noise directly to a clean solution, adding the PDE residual loss only at that final output. The claim is that this decoupling keeps the generative fidelity of the unconditional teacher while substantially improving PDE satisfaction, at a fraction of the sampling cost. A sympathetic reader would care because it offers a simple route to physically consistent generative surrogates for forward simulation, inverse recovery, and reconstruction from partial observations.

What carries the argument

The central object is the distilling student network $d_{\theta'}(\epsilon)$ together with the finite-difference physics residual operator $R(x) = (F[u], B[a])^\top$. The distillation objective couples a regression loss to noise–data pairs produced by the teacher's deterministic ODE with a PDE-residual loss evaluated on the student's one-step output. This moves constraint enforcement from the noisy trajectory to the clean endpoint, which is exactly what removes the Jensen's Gap. The optional refinement step and downstream inference both update the latent noise $\epsilon$ by gradient descent on $R$, which is cheap because the student is one-step.

What would settle it

Train or evaluate PIDDM on a PDE where the loss residual $R$ uses a coarse grid (for example 32$\times$32), then evaluate the same generated samples with a high-resolution or spectral discretization of the same PDE; if the samples' residual on the independent discretization is not correspondingly small, the claim that constraints are enforced on the true sample rather than the discrete proxy would be falsified. A second check: on the correlated mixture-of-Gaussians example, PIDDM should keep the constraint deviation near $2\times10^{-3}$ while matching the target marginal; if either constraint satisfaction or distributional fidelity breaks under different random seeds, the claim of avoiding the trade-off weakens.

Watch

Extended reading notes

Core claim

The central claim is that post-hoc distillation with PDE constraints on the final sample $x = d_{\theta'}(\epsilon)$ avoids the Jensen's Gap that plagues training-time and sampling-time guidance, and yields one-step generation that is both more physics-satisfying and distributionally faithful than PIDM, DiffusionPDE, ECI-sampling, and D-Flow baselines. The method trains a standard teacher diffusion model with a linear flow schedule, generates paired noise–data samples $(\epsilon, x_0)$ by deterministic ODE integration, and trains a student $d_{\theta'}$ with loss $L_{\text{total}} = \mathbb{E}[\|d_{\theta'}(\epsilon)-x_0\|^2] + \lambda_{\text{train}}\|R(x)\|^2$. Because $R$ is evaluated on the actual generated field rather than on a posterior mean estimate, the mismatch $R(\mathbb{E}[x_0|x_t]) \neq \mathbb{E}[R(x_0)|x_t]$ never enters the objective. The paper also demonstrates the gap empirically on a mixture-of-Gaussians example and in training curves, and shows that the distilled student supports forward, inverse, and reconstruction tasks by latent-noise optimization with an observation mask.

Load-bearing premise

The load-bearing premise is that the discretized finite-difference operator $R$ used in training and evaluation is a faithful proxy for the true PDE operator $F$; if the discretization is coarse or biased, the student learns to satisfy the discrete residual rather than the actual physical law, and the reported gains are measured with the same biased ruler.

Editorial extensions

If this is right

  • One-step generation with a single function evaluation achieves PDE residuals lower than all tested guidance-based baselines, and an optional 50-step refinement lowers them further.
  • Distributional fidelity (MMSE and SMSE) does not degrade when the physics loss is added, unlike training-time PIDM whose diffusion loss rises under residual supervision.
  • The same distilled student solves forward, inverse, and partial-observation reconstruction tasks by latent-noise optimization with an observation mask, matching or beating D-Flow at a fraction of its function evaluations.
  • Better noise–data coupling via Rectified Flow or distribution matching distillation further improves the trade-off between sample quality and constraint satisfaction.

Reading between the lines

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

  • Because the finite-difference residual $R$ is both the training objective and the evaluation metric, the reported PDE-error gains are relative to that specific discretization; an independent higher-order or spectral residual check would distinguish true physics satisfaction from fitting the discrete operator.
  • The same post-hoc distillation recipe could be applied to other hard constraints—conservation laws, bounds, symmetries—whenever a teacher diffusion model can produce paired noise–output samples, not just PDE residuals.
  • One-step generation makes latent-space optimization cheap enough for real-time control or data assimilation, a possibility the paper notes but does not develop.
  • If the Jensen's Gap is indeed the dominant source of the fidelity–physics trade-off, then any future method that enforces constraints on the exact final sample should also avoid the trade-off, making the specific distillation mechanism one instance of a more general design principle.
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

4 major / 6 minor

Summary. The paper proposes PIDDM, a post-hoc distillation framework for physics-constrained generation with diffusion models. A teacher diffusion model is trained with the standard (unconstrained) denoising objective; then a student model is trained to map noise to the teacher's deterministic ODE outputs while also minimizing a finite-difference PDE residual on the final sample, thereby avoiding the Jensen's Gap that arises when constraints are applied to the posterior mean. The method also supports optional refinement of the latent noise and is extended to forward/inverse problems and partial-observation reconstruction. Experiments on Darcy, Poisson, Burgers, and several additional PDE benchmarks report improved PDE-residual errors, comparable or better MMSE/SMSE, and lower NFE relative to guidance-based baselines.

Significance. If the empirical claims are validated, the core idea is significant: it decouples physics enforcement from the diffusion trajectory, yields one-step generation, and naturally handles several downstream tasks. The conceptual argument for avoiding Jensen's Gap is clean and the experimental scope is broad, covering eight PDE systems. The paper also includes ablations on key hyperparameters and distillation variants. However, the headline metric is the same finite-difference residual used as the training and refinement objective, so the central claim of improved 'PDE satisfaction' is not yet independently established; the absence of error bars further weakens the comparative claims. The significance is therefore contingent on additional validation.

major comments (4)
  1. [§3.1 Eq. (2); §4.2 Eq. (8); Algorithm 2; §5 PDE Error; Limitations] The reported 'PDE Error' is the squared norm of the same finite-difference residual R that is used as the training loss in Eq. (8) and as the refinement objective in Algorithm 2. Consequently, the student can improve this metric by fitting the discretized stencil rather than the true PDE operator, especially if R is coarse or biased. The Limitations section concedes that constructing a reliable residual operator is challenging with coarse or low-accuracy finite-difference schemes. Because 'significantly improves PDE satisfaction' is the paper's central empirical claim, the authors should add an independent cross-check: for example, evaluate all methods with a higher-order or different discretization, or compare against a numerical solver on a held-out test set, and show that the ranking persists. Without such a check, the headline metric is partly circular.
  2. [Tables 1–4] All quantitative results are single-seed point estimates without error bars or statistical significance tests. Statements such as 'PIDDM significantly improves PDE satisfaction' and 'surpasses all prior methods' are not supported statistically. The authors should repeat experiments over at least three seeds and report mean ± standard deviation, and where possible use paired comparisons on the same latent noise. This is necessary for the reader to assess whether the reported differences (e.g., Darcy MMSE 0.112 vs 0.108 for PIDDM-1 vs Vanilla) are meaningful.
  3. [§5.1, Table 1] The claim that 'PIDDM-1 already surpasses all prior methods that incorporate physical constraints during training or sampling' is contradicted by the table for several fidelity metrics: Poisson SMSE (ECI 0.291 vs PIDDM-1 0.326), Burger SMSE (DiffusionPDE 0.103 vs PIDDM-1 0.133), and Burger MMSE (DiffusionPDE 0.064 vs PIDDM-1 0.152). The claim should be qualified to reference PDE Error or to acknowledge that fidelity is competitive but not uniformly best; the exceptions should be discussed. As written, the text overstates the evidence.
  4. [§3.3, Figure 2] The contribution 'first explicit empirical demonstration of Jensen's Gap' is not established. The MoG experiment compares DPS, with a tuned guidance weight of 0.035, to the ground-truth conditional ODE; the observed velocity errors and constraint violations may reflect DPS's particular guidance approximation rather than Jensen's Gap per se. The training-time experiment in Fig. 2d shows that adding a PIDM-style residual loss increases the diffusion loss, which is a trade-off, but it does not directly measure the discrepancy between R(E[x0|xt]) and E[R(x0)|xt]. A direct measurement of these two quantities for the same score model at multiple noise levels, or a controlled comparison where the only difference is whether the constraint is applied to the posterior mean or the final sample, would substantiate the claim.
minor comments (6)
  1. [§4.2, Algorithm 1] In Algorithm 1, line 5, the residual is written as R(hat(x)), which is ambiguous; it should be R(d_theta'(epsilon)) (or R(x_pred)) to match Eq. (8).
  2. [Appendix B.2] The description of the Helmholtz/Poisson dataset repeats the Darcy Flow setup verbatim ('We adopt the Darcy Flow setup introduced in DiffusionPDE...'); the specific PDE, boundary conditions, and generation process for the Helmholtz/Poisson dataset should be stated explicitly.
  3. [Table 3] The reconstruction Coef MSE values for PIDDM RF-1 and RF-2 (0.913 and 0.954) are an order of magnitude larger than for PIDDM (0.128), yet the text states that RF-2 'achieves the best overall performance across forward and inverse tasks.' Please clarify whether these numbers are correct or transposed, and discuss the reconstruction trade-off.
  4. [Appendix C.3; Tables 1–2] The text says the number of refinement steps N is set to 0 or 50, but Tables 1 and 2 report NFE = 80 for the refinement/optimization settings (0.080 × 10^3). Please define precisely how NFE is counted for Algorithm 2 and Algorithm 3, and reconcile the stated value of N with the reported NFE.
  5. [Contributions list] The bullet 'Theoretically sound' is not supported by a formal theorem or proposition. The conceptual argument against Jensen's Gap is clear, but to avoid overclaiming, either add a formal statement (e.g., a proposition about the training objective or the refinement convergence) or soften the wording to 'conceptually grounded'.
  6. [Throughout] There are several typos and inconsistencies: 'DiuffsionPDE' in §5, 'Diffusin process' in Appendix A.1, and duplicated references [36,37,38] in the bibliography. A careful proofread is needed.

Circularity Check

2 steps flagged · score 6.0 of 10

PDE-error results are partly circular: the reported metric ||R(x)||^2 is exactly the loss minimized in Eq. (8) and in Algorithm 2.

  1. fitted input called prediction [Section 3.1 Eq. (2); Section 4.2 Eq. (8); Section 5 'PDE Error' definition]
    "the physics residual error in often defined by: R(x0)=R((u,a)) := [F[u],B[a]]^T; ... to enforce physical consistency, we evaluate the physics residual error on the output x=dθ′(ε), i.e., ∥R(x)∥2. The overall training objective is: ... +λtrain ∥R(x)∥2; PDE Error quantifies the violation of physical constraints using the physics residual error |R(x)|2."

    Eq. (8) explicitly adds λtrain||R(x)||^2 to the objective minimized during distillation, while Section 5 defines the headline PDE Error as |R(x)|^2. Since R is the same finite-difference operator, the student is trained to minimize exactly the quantity later reported as physical-satisfaction improvement. The PDE Error column is therefore not an independent probe of physical fidelity; it is, up to balancing weight, the training loss itself. The claimed 'significantly improves PDE satisfaction over baselines' is thus partly a restatement of the optimization rather than an external validation. The MMSE/SMSE metrics and qualitative field plots do provide independent content, so the circularity is partial.

  2. fitted input called prediction [Algorithm 2; Section 4.3; Tables 1 and 4 (PIDDM-ref rows)]
    "ϵ←ϵ−ηref∇ϵ∥R(dθ′(ϵ))∥2 ▷PDE refinement step (optional). ... Our optional refinement stage (PIDDM-ref) further reduces both statistical errors and physical PDE residuals."

    PIDDM-ref is produced by gradient descent on ||R(dθ'(ε))||^2, which is exactly the reported PDE Error. The improvement of PIDDM-ref over PIDDM-1 on the PDE-error axis is therefore by construction: it is the result of optimizing the evaluation metric itself. This does not invalidate the refinement as a useful post-processing technique, but it means the reported PDE-error gain cannot serve as independent evidence of better physical fidelity. The same finite-difference caveat applies, and no validation against an independent solver or alternative discretization is provided.

full rationale

The central distillation mechanism is not circular: the student is trained to reproduce teacher outputs via the L_sample term and is evaluated on MMSE/SMSE against ground truth, which are not directly optimized in the same way. The Jensen's Gap demonstration is also an independent empirical comparison. However, the headline 'PDE satisfaction' claim is partially circular. The paper defines the physics residual R(x) via finite differences (Eq. 2), adds lambda_train ||R(x)||^2 to the distillation loss (Eq. 8), refines latent noise by descending the gradient of ||R(d_theta'(epsilon))||^2 (Algorithm 2), and then reports PDE Error as |R(x)|^2. The same residual operator is simultaneously the training objective, the refinement objective, and the evaluation metric. Minimizing a metric and then reporting that metric as evidence of physical fidelity is a by-construction improvement; the magnitude of the gain says little about the true PDE operator unless R is validated independently. The authors' limitation statement admits that a reliable R is assumed, especially with coarse or low-accuracy finite-difference schemes, but no independent solver, finer-grid, or alternative-discretization cross-check is reported. Minor self-citations to prior work by the same authors appear only as motivation for future theoretical work and are not load-bearing for the main empirical claims. Overall, the generative-fidelity claims retain independent content, but the central 'improved PDE satisfaction' claim is substantially entangled with its own loss, so a moderate circularity score is warranted.

Assumptions & free parameters 3 free parameters · 6 assumptions · 0 invented entities

The central method relies on standard diffusion and distillation mathematics, plus a domain assumption that the finite-difference residual is reliable. The only tuned hyperparameters are loss weights and refinement steps; no new physical entities are introduced.

free parameters (3)
  • lambda_train = 10 (most PDEs), 1e6 (Helmholtz and Poisson)
    Loss weight balancing distillation fidelity versus PDE residual; tuned per dataset and the reported PDE-error gains depend on it.
  • lambda_infer = aligned with lambda_train
    Inference-time weighting between observation loss and PDE residual in Algorithm 3.
  • N_f (refinement steps) = 50
    Number of gradient updates on the latent noise in the optional refinement stage; results for PIDDM-ref depend on this choice.
assumptions (6)
  • standard math Tweedie's formula gives a valid posterior-mean estimate from the learned score function (Eq. 3).
    Used to define the Jensen gap and to implement the posterior-mean baselines.
  • standard math Jensen's inequality implies F(E[x0|xt]) is not equal to E[F(x0)|xt] for nonlinear F.
    Core motivation for avoiding posterior-mean constraints; stated with the Jensen gap definition.
  • standard math The probability-flow ODE (Eq. 7) defines a deterministic teacher mapping from noise to data.
    Foundation of the distillation stage; assumes the ODE integration is accurate enough at 100 steps.
  • domain assumption The finite-difference discretized operator R faithfully represents the true PDE operator F.
    Both the training loss and the reported PDE-error metric use this R; if the grid is too coarse or the scheme is biased, results may not reflect true physical fidelity. The authors mention this risk in the limitations.
  • domain assumption A well-trained teacher diffusion model exists and its samples cover the data distribution.
    Distillation cannot exceed the teacher's information; the authors list reliance on a pretrained teacher as a limitation.
  • ad hoc to paper v-prediction with a linear schedule (Eq. 6) produces trajectories smooth enough for one-step student regression.
    A design choice supported by ablations showing VP and sub-VP schedules degrade performance.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Physics-Informed Distillation of Diffusion Models for PDE-Constrained Generation." pith.science (2026). https://pith.science/paper/FZNLCDIC

@misc{pith2026250522391,
  author       = {Pith},
  title        = {Pith review of: Physics-Informed Distillation of Diffusion Models for PDE-Constrained Generation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FZNLCDIC}},
  note         = {Machine review of arXiv:2505.22391}
}
abstract

Modeling physical systems in a generative manner offers several advantages, including the ability to handle partial observations, generate diverse solutions, and address both forward and inverse problems. Recently, diffusion models have gained increasing attention in the modeling of physical systems, particularly those governed by partial differential equations (PDEs). However, diffusion models only access noisy data $\boldsymbol{x}_t$ at intermediate steps, making it infeasible to directly enforce constraints on the clean sample $\boldsymbol{x}_0$ at each noisy level. As a workaround, constraints are typically applied to the expectation of clean samples $\mathbb{E}[\boldsymbol{x}_0|\boldsymbol{x}_t]$, which is estimated using the learned score network. However, imposing PDE constraints on the expectation does not strictly represent the one on the true clean data, known as Jensen's Gap. This gap creates a trade-off: enforcing PDE constraints may come at the cost of reduced accuracy in generative modeling. To address this, we propose a simple yet effective post-hoc distillation approach, where PDE constraints are not injected directly into the diffusion process, but instead enforced during a post-hoc distillation stage. We term our method as Physics-Informed Distillation of Diffusion Models (PIDDM). This distillation not only facilitates single-step generation with improved PDE satisfaction, but also support both forward and inverse problem solving and reconstruction from randomly partial observation. Extensive experiments across various PDE benchmarks demonstrate that PIDDM significantly improves PDE satisfaction over several recent and competitive baselines, such as PIDM, DiffusionPDE, and ECI-sampling, with less computation overhead. Our approach can shed light on more efficient and effective strategies for incorporating physical constraints into diffusion models.

Figures

Figures reproduced from arXiv: 2505.22391 by the authors.

Figure 1
Figure 1. Illustration of physics-constrained diffusion generation and our proposed framework. (a) Existing methods [23, 8, 3, 27] impose PDE losses or guidance on the posterior mean E[x0|xt] in diffusion training and sampling, introducing Jensen’s Gap. (b) We propose to train and sample diffusion model using vanilla methods to generate a noise-image data paired dataset for distillation. (c) Our proposed framework distills th… view at source ↗
Figure 2
Figure 2. Empirical illustration of the Jensen’s Gap in physics-constrained diffusion models. (a) Absolute velocity error and angular discrepancy (1 − cos(θ)) between Diffusion Posterior Sampling (DPS) and the ground-truth conditional ODE velocity on the MoG dataset. (b) and (c) Histograms comparing the first (unconstrained) and second (hard-constrained) dimensions of DPS-sampled MoG data against the ground truth MoG. (d) Tra… view at source ↗
Figure 3
Figure 3. Ablation studies on the effect of several factors on the performance of PIDDM on Darcy [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: compares the predicted Darcy pressure fields and their corresponding data- and PDE-error maps for each baseline and for our PIDDM. DiffuionPDE, and ECI reproduce the coarse flow pattern but exhibit large point-wise errors and pronounced residual bands. In contrast, PID…
Figure 5
Figure 5. Figure 5: Constraint satisfaction on correlated MoG. Comparison of generated samples using DPS, [PITH_FULL_IMAGE:figures/full_fig_p023_5.png]

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

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

  1. Lantern: Conflict-Aware Gradient Blending for Physics-Guided Diffusion Models in Calorimeter Simulation

    cs.LG 2026-07 conditional novelty 6.5 of 10

    GradBlend anchors diffusion updates to denoising while admitting physics auxiliaries, improving calorimeter shower FPD and CFD where PCGrad, GradNorm, IMTL-G, and ConFIG inflate FPD by 2–100×.

Reference graph

Works this paper leans on

67 extracted references · 18 canonical work pages · cited by 1 Pith paper

  1. [1]

    {TensorFlow}: a system for {Large-Scale} machine learning

    Martín Abadi, Paul Barham, Jianmin Chen, Zhifeng Chen, Andy Davis, Jeffrey Dean, Matthieu Devin, Sanjay Ghemawat, Geoffrey Irving, Michael Isard, et al. {TensorFlow}: a system for {Large-Scale} machine learning. In12th USENIX symposium on operating systems design and implementation (OSDI 16), pages 265–283, 2016

    2016

  2. [2]

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

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

    1982

  3. [3]

    Physics-informed diffusion models

    Jan-Hendrik Bastek, WaiChing Sun, and Dennis Kochmann. Physics-informed diffusion models. InThe Thirteenth International Conference on Learning Representations, 2025

    2025

  4. [4]

    D-flow: Differentiating through flows for controlled generation.arXiv preprint arXiv:2402.14017, 2024

    Heli Ben-Hamu, Omri Puny, Itai Gat, Brian Karrer, Uriel Singer, and Yaron Lipman. D-flow: Differentiating through flows for controlled generation.arXiv preprint arXiv:2402.14017, 2024. 11

    arXiv 2024

  5. [5]

    Tract: Denoising diffusion models with transitive closure time-distillation.arXiv preprint arXiv:2303.04248, 2023

    David Berthelot, Arnaud Autef, Jierui Lin, Dian Ang Yap, Shuangfei Zhai, Siyuan Hu, Daniel Zheng, Walter Talbott, and Eric Gu. Tract: Denoising diffusion models with transitive closure time-distillation.arXiv preprint arXiv:2303.04248, 2023

    arXiv 2023

  6. [6]

    Find: Fine-tuning initial noise distribution with policy optimization for diffusion models

    Changgu Chen, Libing Yang, Xiaoyan Yang, Lianggangxu Chen, Gaoqi He, Changbo Wang, and Yang Li. Find: Fine-tuning initial noise distribution with policy optimization for diffusion models. InProceedings of the 32nd ACM International Conference on Multimedia, pages 6735–6744, 2024

    work page 2024

  7. [7]

    Masked autoencoders are effective tokenizers for diffusion models.arXiv preprint arXiv:2502.03444, 2025

    Hao Chen, Yujin Han, Fangyi Chen, Xiang Li, Yidong Wang, Jindong Wang, Ze Wang, Zicheng Liu, Difan Zou, and Bhiksha Raj. Masked autoencoders are effective tokenizers for diffusion models.arXiv preprint arXiv:2502.03444, 2025

    arXiv 2025

  8. [8]

    Maddix, Abdul Fatir Ansari, Andrew Stuart, Michael W

    Chaoran Cheng, Boran Han, Danielle C. Maddix, Abdul Fatir Ansari, Andrew Stuart, Michael W. Mahoney, and Bernie Wang. Gradient-free generation for hard-constrained systems. InThe Thirteenth International Conference on Learning Representations, 2025

    work page 2025

Show all 67 references
  1. [9]

    Diffu- sion posterior sampling for general noisy inverse problems.arXiv preprint arXiv:2209.14687, 2022

    Hyungjin Chung, Jeongsol Kim, Michael T Mccann, Marc L Klasky, and Jong Chul Ye. Diffu- sion posterior sampling for general noisy inverse problems.arXiv preprint arXiv:2209.14687, 2022

    2022 arXiv

  2. [10]

    Oxford University Press, 2 edition, 1975

    John Crank.The Mathematics of Diffusion. Oxford University Press, 2 edition, 1975

    1975

  3. [11]

    Davidson.Turbulence: An Introduction for Scientists and Engineers

    Peter A. Davidson.Turbulence: An Introduction for Scientists and Engineers. Oxford University Press, 2015

    2015

  4. [12]

    The helmholtz machine.Neural computation, 7(5):889–904, 1995

    Peter Dayan, Geoffrey E Hinton, Radford M Neal, and Richard S Zemel. The helmholtz machine.Neural computation, 7(5):889–904, 1995

    1995

  5. [13]

    Tweedie’s formula and selection bias.Journal of the American Statistical Association, 106(496):1602–1614, 2011

    Bradley Efron. Tweedie’s formula and selection bias.Journal of the American Statistical Association, 106(496):1602–1614, 2011

    2011

  6. [14]

    Scaling rectified flow transform- ers for high-resolution image synthesis

    Patrick Esser, Sumith Kulal, Andreas Blattmann, Rahim Entezari, Jonas Müller, Harry Saini, Yam Levi, Dominik Lorenz, Axel Sauer, Frederic Boesel, et al. Scaling rectified flow transform- ers for high-resolution image synthesis. InForty-first International Conference on Machine...

    2024

  7. [15]

    Bounds on the jensen gap, and implications for mean-concentrated distributions.arXiv preprint arXiv:1712.05267, 2017

    Xiang Gao, Meera Sitharam, and Adrian E Roitberg. Bounds on the jensen gap, and implications for mean-concentrated distributions.arXiv preprint arXiv:1712.05267, 2017

    2017 arXiv

  8. [16]

    Initno: Boosting text-to-image diffusion models via initial noise optimization

    Xiefan Guo, Jinlin Liu, Miaomiao Cui, Jiankai Li, Hongyu Yang, and Di Huang. Initno: Boosting text-to-image diffusion models via initial noise optimization. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 9380–9389, 2024

    2024

  9. [17]

    On the feature learning in diffusion models

    Andi Han, Wei Huang, Yuan Cao, and Difan Zou. On the feature learning in diffusion models. arXiv preprint arXiv:2412.01021, 2024

    2024 arXiv

  10. [18]

    Can diffusion models learn hidden inter-feature rules behind images?arXiv preprint arXiv:2502.04725, 2025

    Yujin Han, Andi Han, Wei Huang, Chaochao Lu, and Difan Zou. Can diffusion models learn hidden inter-feature rules behind images?arXiv preprint arXiv:2502.04725, 2025

    2025 arXiv

  11. [19]

    Learning physical models that can respect conservation laws

    Derek Hansen, Danielle C Maddix, Shima Alizadeh, Gaurav Gupta, and Michael W Mahoney. Learning physical models that can respect conservation laws. InInternational Conference on Machine Learning, pages 12469–12510. PMLR, 2023

    2023

  12. [20]

    Reducing the dimensionality of data with neural networks.science, 313(5786):504–507, 2006

    Geoffrey E Hinton and Ruslan R Salakhutdinov. Reducing the dimensionality of data with neural networks.science, 313(5786):504–507, 2006

    2006

  13. [21]

    Denoising diffusion probabilistic models.Advances in neural information processing systems, 33:6840–6851, 2020

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffusion probabilistic models.Advances in neural information processing systems, 33:6840–6851, 2020

    2020

  14. [22]

    Classifier-free diffusion guidance, 2022

    Jonathan Ho and Tim Salimans. Classifier-free diffusion guidance, 2022

    2022

  15. [23]

    DiffusionPDE: Generative PDE-solving under partial observation

    Jiahe Huang, Guandao Yang, Zichen Wang, and Jeong Joon Park. DiffusionPDE: Generative PDE-solving under partial observation. InThe Thirty-eighth Annual Conference on Neural Information Processing Systems, 2024. 12

    2024

  16. [24]

    Courier Corporation, 2003

    Thomas JR Hughes.The finite element method: linear static and dynamic finite element analysis. Courier Corporation, 2003

    2003

  17. [25]

    Incropera, David P

    Frank P. Incropera, David P. DeWitt, Theodore L. Bergman, and Adrienne S. Lavine.Funda- mentals of Heat and Mass Transfer. John Wiley & Sons, 7 edition, 2011

    2011

  18. [26]

    John Wiley & Sons, 3 edition, 1998

    John David Jackson.Classical Electrodynamics. John Wiley & Sons, 3 edition, 1998

    1998

  19. [27]

    Cocogen: Physically-consistent and conditioned score-based generative models for forward and inverse problems, 2024

    Christian Jacobsen, Yilin Zhuang, and Karthik Duraisamy. Cocogen: Physically-consistent and conditioned score-based generative models for forward and inverse problems, 2024

    2024

  20. [28]

    Elucidating the design space of diffusion-based generative models.Advances in neural information processing systems, 35:26565–26577, 2022

    Tero Karras, Miika Aittala, Timo Aila, and Samuli Laine. Elucidating the design space of diffusion-based generative models.Advances in neural information processing systems, 35:26565–26577, 2022

    2022

  21. [29]

    Functional flow matching.arXiv preprint arXiv:2305.17209, 2023

    Gavin Kerrigan, Giosue Migliorini, and Padhraic Smyth. Functional flow matching.arXiv preprint arXiv:2305.17209, 2023

    2023 arXiv

  22. [30]

    Adam: A method for stochastic optimization.arXiv preprint arXiv:1412.6980, 2014

    Diederik P Kingma and Jimmy Ba. Adam: A method for stochastic optimization.arXiv preprint arXiv:1412.6980, 2014

    2014 arXiv

  23. [31]

    Auto-encoding variational bayes, 2013

    Diederik P Kingma, Max Welling, et al. Auto-encoding variational bayes, 2013

    2013

  24. [32]

    SIAM, 2007

    Randall J LeVeque.Finite difference methods for ordinary and partial differential equations: steady-state and time-dependent problems. SIAM, 2007

    2007

  25. [33]

    Randall J LeVeque and Randall J Leveque.Numerical methods for conservation laws, volume

  26. [34]

    Fourier neural operator for parametric partial differen- tial equations.arXiv preprint arXiv:2010.08895, 2020

    Zongyi Li, Nikola Kovachki, Kamyar Azizzadenesheli, Burigede Liu, Kaushik Bhattacharya, Andrew Stuart, and Anima Anandkumar. Fourier neural operator for parametric partial differen- tial equations.arXiv preprint arXiv:2010.08895, 2020

    2010 arXiv

  27. [35]

    Physics-informed neural operator for learning partial differential equations.ACM/JMS Journal of Data Science, 1(3):1–27, 2024

    Zongyi Li, Hongkai Zheng, Nikola Kovachki, David Jin, Haoxuan Chen, Burigede Liu, Kamyar Azizzadenesheli, and Anima Anandkumar. Physics-informed neural operator for learning partial differential equations.ACM/JMS Journal of Data Science, 1(3):1–27, 2024

    2024

  28. [36]

    Yaron Lipman, Ricky T. Q. Chen, Heli Ben-Hamu, Maximilian Nickel, and Matthew Le. Flow matching for generative modeling. InThe Eleventh International Conference on Learning Representations, 2023

    2023

  29. [38]

    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. InThe Eleventh International Conference on Learning Representations, 2023

    2023

  30. [39]

    Instaflow: One step is enough for high-quality diffusion-based text-to-image generation

    Xingchao Liu, Xiwen Zhang, Jianzhu Ma, Jian Peng, et al. Instaflow: One step is enough for high-quality diffusion-based text-to-image generation. InThe Twelfth International Conference on Learning Representations, 2023

    2023

  31. [40]

    Dpm-solver: A fast ode solver for diffusion probabilistic model sampling in around 10 steps.Advances in Neural Information Processing Systems, 35:5775–5787, 2022

    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, 35:5775–5787, 2022

    2022

  32. [41]

    Dpm- solver++: Fast solver for guided sampling of diffusion probabilistic models.arXiv preprint arXiv:2211.01095, 2022

    Cheng Lu, Yuhao Zhou, Fan Bao, Jianfei Chen, Chongxuan Li, and Jun Zhu. Dpm- solver++: Fast solver for guided sampling of diffusion probabilistic models.arXiv preprint arXiv:2211.01095, 2022

    2022 arXiv

  33. [42]

    Deeponet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operators

    Lu Lu, Pengzhan Jin, and George Em Karniadakis. Deeponet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operators. arXiv preprint arXiv:1910.03193, 2019. 13

    1910 arXiv

  34. [43]

    Knowledge distillation in iterative generative models for improved sampling speed.arXiv preprint arXiv:2101.02388, 2021

    Eric Luhman and Troy Luhman. Knowledge distillation in iterative generative models for improved sampling speed.arXiv preprint arXiv:2101.02388, 2021

    2021 arXiv

  35. [44]

    The lottery ticket hypothesis in denoising: Towards semantic-driven initialization

    Jiafeng Mao, Xueting Wang, and Kiyoharu Aizawa. The lottery ticket hypothesis in denoising: Towards semantic-driven initialization. InEuropean Conference on Computer Vision, pages 93–109. Springer, 2024

    2024

  36. [45]

    Interacting particle solutions of fokker–planck equations through gradient–log–density estimation.Entropy, 22(8):802, 2020

    Dimitra Maoutsa, Sebastian Reich, and Manfred Opper. Interacting particle solutions of fokker–planck equations through gradient–log–density estimation.Entropy, 22(8):802, 2020

    2020

  37. [46]

    Using uncertainty quantification to characterize and improve out-of-domain learning for pdes.arXiv preprint arXiv:2403.10642, 2024

    S Chandra Mouli, Danielle C Maddix, Shima Alizadeh, Gaurav Gupta, Andrew Stuart, Michael W Mahoney, and Yuyang Wang. Using uncertainty quantification to characterize and improve out-of-domain learning for pdes.arXiv preprint arXiv:2403.10642, 2024

    2024 arXiv

  38. [47]

    Improved denoising diffusion probabilistic models

    Alexander Quinn Nichol and Prafulla Dhariwal. Improved denoising diffusion probabilistic models. InInternational conference on machine learning, pages 8162–8171. PMLR, 2021

    2021

  39. [48]

    Pytorch: An imperative style, high-performance deep learning library.arXiv preprint arXiv:1912.01703, 2019

    A Paszke. Pytorch: An imperative style, high-performance deep learning library.arXiv preprint arXiv:1912.01703, 2019

    1912 arXiv

  40. [49]

    Maziar Raissi, Paris Perdikaris, and George E Karniadakis. Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations.Journal of Computational physics, 378:686–707, 2019

    2019

  41. [50]

    Stochastic backpropagation and approximate inference in deep generative models

    Danilo Jimenez Rezende, Shakir Mohamed, and Daan Wierstra. Stochastic backpropagation and approximate inference in deep generative models. InInternational conference on machine learning, pages 1278–1286. PMLR, 2014

    2014

  42. [51]

    High- resolution image synthesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Björn Ommer. High- resolution image synthesis with latent diffusion models. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10684–10695, 2022

    2022

  43. [52]

    Guiding continuous operator learning through physics-based boundary constraints.arXiv preprint arXiv:2212.07477, 2022

    Nadim Saad, Gaurav Gupta, Shima Alizadeh, and Danielle C Maddix. Guiding continuous operator learning through physics-based boundary constraints.arXiv preprint arXiv:2212.07477, 2022

    2022 arXiv

  44. [53]

    Parallel sampling of diffusion models.Advances in Neural Information Processing Systems, 36, 2024

    Andy Shih, Suneel Belkhale, Stefano Ermon, Dorsa Sadigh, and Nima Anari. Parallel sampling of diffusion models.Advances in Neural Information Processing Systems, 36, 2024

    2024

  45. [54]

    A physics-informed diffusion model for high- fidelity flow field reconstruction.Journal of Computational Physics, 478:111972, 2023

    Dule Shu, Zijie Li, and Amir Barati Farimani. A physics-informed diffusion model for high- fidelity flow field reconstruction.Journal of Computational Physics, 478:111972, 2023

    2023

  46. [55]

    Oxford university press, 1985

    Gordon D Smith.Numerical solution of partial differential equations: finite difference methods. Oxford university press, 1985

    1985

  47. [56]

    Denoising diffusion implicit models

    Jiaming Song, Chenlin Meng, and Stefano Ermon. Denoising diffusion implicit models. In International Conference on Learning Representations, 2021

    2021

  48. [57]

    Consistency models, 2023

    Yang Song, Prafulla Dhariwal, Mark Chen, and Ilya Sutskever. Consistency models, 2023

    2023

  49. [58]

    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, 2020

    2020

  50. [59]

    Timoshenko and James N

    Stephen P. Timoshenko and James N. Goodier.Theory of Elasticity. McGraw-Hill, 3 edition, 1970

    1970

  51. [60]

    Attention is all you need.Advances in neural information processing systems, 30, 2017

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need.Advances in neural information processing systems, 30, 2017

    2017

  52. [61]

    Noise re-sampling for high fidelity image generation, 2025

    Hao Wang, Weihua Chen, Chenming Li, Wenjian Huang, Jingkai Zhou, Fan Wang, and Jianguo Zhang. Noise re-sampling for high fidelity image generation, 2025. 14

    2025

  53. [62]

    The silent prompt: Initial noise as implicit guidance for goal-driven image generation.arXiv preprint arXiv:2412.05101, 2024

    Ruoyu Wang, Huayang Huang, Ye Zhu, Olga Russakovsky, and Yu Wu. The silent prompt: Initial noise as implicit guidance for goal-driven image generation.arXiv preprint arXiv:2412.05101, 2024

    2024 arXiv

  54. [63]

    One-step diffusion with distribution matching distillation

    Tianwei Yin, Michaël Gharbi, Richard Zhang, Eli Shechtman, Fredo Durand, William T Freeman, and Taesung Park. One-step diffusion with distribution matching distillation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 6613–6623, 2024

    2024

  55. [64]

    Representation alignment for generation: Training diffusion transformers is easier than you think.arXiv preprint arXiv:2410.06940, 2024

    Sihyun Yu, Sangkyung Kwak, Huiwon Jang, Jongheon Jeong, Jonathan Huang, Jinwoo Shin, and Saining Xie. Representation alignment for generation: Training diffusion transformers is easier than you think.arXiv preprint arXiv:2410.06940, 2024

    arXiv

  56. [65]

    Fast sampling of diffusion models via operator learning

    Hongkai Zheng, Weili Nie, Arash Vahdat, Kamyar Azizzadenesheli, and Anima Anandkumar. Fast sampling of diffusion models via operator learning. InInternational conference on machine learning, pages 42390–42402. PMLR, 2023

    2023

  57. [66]

    Fast ode-based sampling for diffusion models in around 5 steps

    Zhenyu Zhou, Defang Chen, Can Wang, and Chun Chen. Fast ode-based sampling for diffusion models in around 5 steps. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 7777–7786, 2024

    2024

  58. [67]

    Golden noise for diffusion models: A learning framework.arXiv preprint arXiv:2411.09502, 2024

    Zikai Zhou, Shitong Shao, Lichen Bai, Zhiqiang Xu, Bo Han, and Zeke Xie. Golden noise for diffusion models: A learning framework.arXiv preprint arXiv:2411.09502, 2024. 15 A Mixture-of-Gaussians (MoG) Dataset To study the sampling-time behavior of constrained diffusion models, ...

    2024 arXiv

  59. [68]

    The weighting λtrain is aligned with our setup across datasets

    and introduce an additional PDE residual loss evaluated on the final sample. The weighting λtrain is aligned with our setup across datasets. Specifically, the implementation follows the D-Flow setup in ECI-sampling [ 8]: we discretize the sampling trajectory into 100 denoising...

    1950

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.