Pith. sign in

REVIEW 5 major objections 5 minor 2 cited by

Incorporating Pre-trained Diffusion Models in Solving the Schr\"odinger Bridge Problem

T0 review · 5 major / 5 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read The paper shows that the Diffusion Schrödinger Bridge objective can be rewritten as mean-, terminus-, or flow-matching, allowing pre-trained diffusion models to initialize bridge training and surpass the pre-trained base model.

desk verdict Useful practical recipe for SB training via diffusion-style losses and SGM initialization, with solid empirical gains, but the claimed DSB equivalence rests on unquantified approximations and missing proofs. read the letter →

arxiv 2508.18095 v1 pith:FRJ3TTWT submitted 2025-08-25 cs.CV cs.LG

Zhicong Tang , Tiankai Hang , Shuyang Gu , Dong Chen , Baining Guo This is my paper
classification cs.CVcs.LG
keywords Schrödingerbridgediffusionmodelsscore-basedgenerativeiterativeproportionalfittingflowmatchingimage-to-imagetranslationpre-trainedinitializationmodeling
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 tries to establish that the slow, unstable iterative training of Schrödinger-bridge generative models can be fixed by rewriting the Diffusion Schrödinger Bridge objective in three equivalent forms—mean-matching, terminus-matching, and flow-matching—so that the network's job is the same kind of simple prediction that diffusion models already make. In these forms, an off-the-shelf pre-trained diffusion model can be dropped in as the initial bridge network, and the iterative bridge fine-tuning then improves on what the diffusion model alone can do. The authors argue that pre-trained initialization is only compatible with the reparameterized objectives, not with the original DSB form, and support this with experiments where initialized IPFM models reduce AFHQ cat-to-dog FID from 15.46 to 9.96. If the equivalence arguments hold, the result unifies diffusion models and Schrödinger bridges and gives practitioners a concrete recipe: start from a pre-trained diffusion model, fine-tune with a bridge objective, and obtain a transport between arbitrary distributions that the diffusion prior alone cannot express.

What carries the argument

The load-bearing objects are the three reparameterized objectives IPMM, IPTM, and IPFM, together with the linear interpolation identity xk−1 = xk + (1/N)(x0 − x1), which translates a pre-trained diffusion model's prediction (the vector x0 − x1) into a bridge step. Equations (11) and (13) give the exact algebraic maps between the new prediction targets and the original DSB networks B and F, while Propositions 1 and 2 supply the Gaussian and small-step approximations that let the new objectives inherit the DSB fixed point. The shared noise schedule between the pre-trained SGM and the bridge is what makes the initialization plug-and-play.

What would settle it

Run the proposed IPFM training for a Gaussian pair with a known closed-form Schrödinger bridge, using a coarse discretization (small N) and a strongly nonlinear initial drift; if the recovered joint coupling differs measurably from the analytic bridge beyond numerical tolerance, the claimed equivalence of the reparameterized objectives to DSB fails in that regime. A cheaper check is to Monte Carlo estimate the conditional posterior p(xk|xk+1,x0) and compare its mean to the linear expression in Equation (16).

Watch

Extended reading notes

Core claim

The central claim, stated on the paper's own terms, is that the DSB objective in Equation (8) is not the only practical objective for training Schrödinger-bridge networks: it can be reparameterized into IPMM (predict the immediate next state), IPTM (predict the trajectory terminus x0 or xN), and IPFM (predict the vector from the current state to the terminus). Proposition 1 states these losses are approximately equal to the DSB losses under Gaussian transitions, and Proposition 2 states that, under mild assumptions, the conditional posterior of intermediate bridge states is Gaussian with the linear means shown in Equation (16). Because the resulting prediction targets match those of standard

Load-bearing premise

The bridge objectives inherit the Schrödinger bridge guarantee only if the intermediate bridge transitions are well approximated by Gaussians with a linear conditional mean and the timesteps are small; those approximations (Propositions 1 and 2) are asserted rather than quantified, with proofs deferred to appendices that are not present.

Editorial extensions

If this is right

  • Pre-trained SGMs become drop-in initializers for bridge models, so Schrödinger-bridge training no longer starts from random networks; the bridge fine-tuning then improves on the SGM's own FID.
  • Each prediction target costs one network forward evaluation instead of two in the original DSB objective, roughly halving the per-epoch training cost.
  • Training stability improves because the networks regress toward endpoints or directions that are real samples, not toward the output of another network that is itself changing.
  • The unified objective form means SGM infrastructure—architectures, noise schedules, solvers—can be reused directly in bridge models.
  • The method works beyond unpaired translation, also improving class-conditional and text-conditional generation over the pre-trained SGM on ImageNet and CUB-200.

Reading between the lines

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

  • An unstated corollary of the equivalence is that in the small-timestep limit IPFM behaves like a bridge-space flow matching; one could test whether learned bridge trajectories become approximately straight and can be integrated with fast ODE solvers.
  • The initialization recipe inherits the pre-trained model's noise schedule, so a testable prediction is that mismatched schedules between SGM and bridge will erode or reverse the gains.
  • If Proposition 2's Gaussian posterior approximation fails on nonlinear data manifolds at coarse discretizations, the method's benefits may shrink at small N; an explicit stress test of very few timesteps would bound that regime.
  • The paper's evidence that original DSB is harmed by SGM initialization suggests that other non-reparameterized bridge solvers would need a similar algebraic rewrite before they can exploit the existing base of pre-trained diffusion models.
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

5 major / 5 minor

Summary. The paper proposes three reparameterized training objectives for diffusion Schrödinger bridge (DSB) models — IPMM, IPTM, and IPFM — which are claimed to be equivalent to the original DSB objective (Eq. 8) and therefore to inherit DSB's convergence to the Schrödinger bridge solution. The authors further propose initializing SB networks with pre-trained score-based generative models (SGMs), arguing that such initialization accelerates and stabilizes training. The empirical section reports improved FID scores on CelebA, AFHQ, Horse2Zebra, Selfie2Anime, ImageNet, and CUB-200, in some cases substantially outperforming DSB, IPML, DSBM, and other baselines. The central theoretical claim is that the three reparameterizations are equivalent to the DSB objective; this equivalence is asserted through approximate statements (Eqs. 14–16) with proofs deferred to appendices B and C, which are absent from the submitted manuscript.

Significance. If the equivalence and convergence claims were rigorously established, the paper would make a useful practical contribution by connecting SGM training recipes with Schrödinger bridge solvers and by showing that SGM initialization improves SB training. The empirical results are extensive and, assuming they are reproducible, demonstrate meaningful FID improvements and a reduced number of forward evaluations. However, the paper's main theoretical guarantee — inheritance of DSB/IPF convergence — is currently unsupported because the equivalence proofs are missing and the key relations are approximate with unquantified error. The strength of the paper is therefore primarily empirical at this stage. The authors are honest about the approximations and explicitly note that proofs are deferred, but a journal submission cannot simply omit the appendices in which those proofs are claimed to live.

major comments (5)
  1. [§4.4, Proposition 1 (Eq. 14)] The equivalence L'_B ≈ L_B and L'_F ≈ L_F is asserted with 'similar to Equation (4)' and the proof is deferred to Appendix B, which is missing. This is a load-bearing claim: if the two objectives have different minimizers, then IPMM is not equivalent to DSB and the inherited convergence to the Schrödinger bridge does not follow. Under the stated Gaussian transition assumption, L' trains the network to predict the conditional mean of the previous state, while L trains it to predict x_{k+1} + F(x_k) − F(x_{k+1}); these targets agree only up to an error that depends on the step size and the smoothness/Lipschitz constant of F. Please provide a precise statement, a proof, and an explicit error bound in terms of γ_k and the network regularity, or weaken the claim accordingly.
  2. [§4.4, Proposition 2 (Eqs. 15–16)] The Gaussian conditional posterior with linear mean μ = x_{k+1} + γ_{k+1}/barγ_{k+1} (x_0 − x_{k+1}) is exact only for linear-Gaussian forward transitions with known x_0. The 'mild assumptions' are never stated, and the actual DSB/SB models use nonlinear neural-network drifts F(x) and B(x). For nonlinear drift, the posterior is generally non-Gaussian and its mean is not the stated linear expression. Consequently the IPTM and IPFM targets (x_0 and (x_0 − x_{k+1})/barγ) are biased estimates of the true conditional expectations, and the claimed equivalence of IPTM/IPFM to IPMM/DSB fails in the practical regime. This is directly load-bearing for the paper's theoretical contribution. Please state the exact assumptions under which Eqs. (15)–(16) hold, prove the approximation with a quantified error, or reformulate the objectives as approximate/variational objectives whose convergence propertie
  3. [§5.1, Eq. (17)] The convergence analysis invokes the assumption that every epoch achieves the completely converged state, i.e. q^n_{i|i+1} = p^n_{i|i+1} for all i. Under that assumption, q^n_0 = p^n_0 = p_data, so the left-hand side π^{2n+1}_0 − p_data is zero. The right-hand side is an integral of (pprior − p^n_N) over a product of conditional densities, which is generally nonzero. The displayed identity therefore either is inconsistent or requires additional hidden conditions. The qualitative conclusion that training relies on the previous epoch may be true heuristically, but the formal derivation as written does not support it. Please correct the derivation or explicitly label this argument as heuristic.
  4. [§6, Tables 3–6] The reported FID numbers are single runs with no error bars, multiple seeds, or statistical significance testing. For example, in Table 3, cat→dog goes from 11.54 (IPTM) to 9.96 (IPFM), and in Table 4 the differences between IPTM and IPFM on several tasks are of similar magnitude to typical run-to-run variance. The empirical claims would be considerably stronger with seed variation and standard errors. Additionally, the 'pre-trained SGMs' used in the experiments are models trained by the authors on each dataset (Section 6.4), not off-the-shelf checkpoints such as Stable Diffusion. The abstract and introduction imply a plug-and-play use of existing SGMs; please clarify what is actually pre-trained and what is trained in-house.
  5. [Appendix references] The text explicitly states 'We leave the detailed proof in Appendix B and Appendix C' (§4.4) and 'We leave the details in the Appendix A' (§5.2), but none of these appendices appears in the submitted manuscript. A journal submission cannot rely on material that is not present for central proofs. Please include the appendices or remove the references. This is not a stylistic point: the absent appendices contain the proof of the paper's main theoretical claim.
minor comments (5)
  1. [Throughout] Several typos remain: 'pretained' in Table 1 caption; 'Fid' in Table 6; 'reperameterized' in §4.4; 'the the' in §5.1. A careful proofread is needed.
  2. [§2.1, Eq. (4)] The reverse-time approximation invokes pk ≈ pk+1 and a Taylor expansion of log pk+1, but the required regularity and step-size conditions are not stated. Please add the standard assumptions (smooth score, small γ) so that Eq. (4) is formally justified.
  3. [§3.2, Eq. (7)] The notation in Eq. (7) is confusing: the KL divergence is written between conditional densities πk|k+1 and π2n_{k|k+1}, but the optimization is over π ∈ PN+1 with a marginal constraint. Clarify the exact functional being minimized and the relationship to the IPF updates in Eq. (6).
  4. [§5.2, Eq. (19)–(20)] The derivation of the initialization x_{k−1} = x_k + (1/N)(x_0 − x_1) assumes a specific linear interpolation schedule. For other schedules (VP, VE) the recalibration is only mentioned in passing; please provide the explicit formulas or a reference, since the paper claims generality to 'other noise schedules'.
  5. [§6.4] Figure 6 and the discussion of γ schedule are qualitative. Please provide quantitative results (e.g., FID vs. alignment metric) for different γ schedules, or clearly state that this is an observation rather than a systematic study.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the IPMM/IPTM/IPFM objectives are reparameterizations or approximations of the DSB objective, not fitted inputs renamed as predictions; missing appendices are a rigor gap, not circularity.

full rationale

IPMM (Eq. 9) is a direct regression of the backward network to the previous state x_k, and Proposition 1 equates it to DSB's Eq. (8) via the standard small-step approximation of Eq. (4) — an approximation, not an identity forced by definition. IPTM and IPFM are connected to IPMM by exact algebraic reparameterizations (Eqs. 11 and 13), and Proposition 2 supplies the additional Gaussian-conditional approximation whose proof is deferred ('We leave the detailed proof in Appendix B and Appendix C'); the appendices are absent from the submitted v1. That is an unquantified support gap, not circularity. The claimed convergence is inherited from external DSB/IPF theory rather than from the authors' own fitted values, and the training targets (x_0, x_N, or their differences) are ground-truth trajectory quantities, not outputs of the networks being trained. The only self-citation, [44] in the introductory remark '3D content [21, 44]', is contextual and not load-bearing. The paper's weakness is the loose '≈' in Propositions 1 and 2 and the missing proofs, which are correctness/rigor concerns outside the circularity rubric.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central theoretical claim rests on unquantified approximate equalities and an unspecified 'mild assumptions' posterior. The only hand-set numeric quantities are hyperparameters (noise schedule, N, epochs). No new physical or mathematical entities are introduced.

free parameters (3)
  • Noise schedule gamma_k = Symmetric schedule from gamma_min to gamma_max, values per dataset not fully stated
    Chosen by hand following [30]; the paper shows the schedule controls the alignment-quality trade-off (Figure 6), so results depend on this choice.
  • Number of timesteps N = 50 for CelebA; not clearly reported for other tasks
    Discretization hyperparameter chosen by hand; the approximation quality of Propositions 1 and 2 depends on step size.
  • Training epochs L = 2/4/6/8/12 depending on task
    Early stopping per dataset chosen by FID; affects the reported improvements.
assumptions (4)
  • domain assumption DSB iterative training converges to the Schrödinger bridge solution via IPF
    Relied on in Section 4.4 to transfer convergence from DSB to the new objectives; proved in cited literature [10, 13, 26].
  • domain assumption Gaussian transition kernels and small-step approximation of the reverse process (Eq. 4)
    Standard SGM approximation used in Proposition 1 to identify the DSB target with x_k and to justify IPMM.
  • ad hoc to paper Posterior of x_k given x_{k+1}, x0 is Gaussian with the linear mean in Eqs. 15 and 16
    Proposition 2 asserts this under unspecified 'mild assumptions'; it is needed so IPTM/IPFM targets are valid conditional means. It is not exact for general nonlinear drift.
  • domain assumption Pre-trained SGM and SB share the same noise schedule and input distribution
    The initialization formula in Section 5.2 requires that x_k distributions match; the authors state this but do not show how to adapt off-the-shelf models with different schedules.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Incorporating Pre-trained Diffusion Models in Solving the Schr\"odinger Bridge Problem." pith.science (2026). https://pith.science/paper/FRJ3TTWT

@misc{pith2026250818095,
  author       = {Pith},
  title        = {Pith review of: Incorporating Pre-trained Diffusion Models in Solving the Schr\"odinger Bridge Problem},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FRJ3TTWT}},
  note         = {Machine review of arXiv:2508.18095}
}
read the original abstract

This paper aims to unify Score-based Generative Models (SGMs), also known as Diffusion models, and the Schr\"odinger Bridge (SB) problem through three reparameterization techniques: Iterative Proportional Mean-Matching (IPMM), Iterative Proportional Terminus-Matching (IPTM), and Iterative Proportional Flow-Matching (IPFM). These techniques significantly accelerate and stabilize the training of SB-based models. Furthermore, the paper introduces novel initialization strategies that use pre-trained SGMs to effectively train SB-based models. By using SGMs as initialization, we leverage the advantages of both SB-based models and SGMs, ensuring efficient training of SB-based models and further improving the performance of SGMs. Extensive experiments demonstrate the significant effectiveness and improvements of the proposed methods. We believe this work contributes to and paves the way for future research on generative models.

Discussion (0). Sign in to comment.

Forward citations

Cited by 2 Pith papers

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

  1. Deterministic Decomposition of Stochastic Generative Dynamics

    cs.LG 2026-05 unverdicted novelty 5.0 of 10

    Stochastic generative dynamics admit a transport-osmotic decomposition of the deterministic field, supporting Bridge Matching for interpretable and tunable generation.

  2. Deterministic Decomposition of Stochastic Generative Dynamics

    cs.LG 2026-05 unverdicted novelty 5.0 of 10

    Stochastic generative dynamics are decomposed into transport and osmotic parts via b_t = u_t + d_t, with Bridge Matching proposed to learn the components for controllable sampling.

Reference graph

Works this paper leans on

50 extracted references · 41 canonical work pages · cited by 1 Pith paper

  1. [1]

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

    work page 1982

  2. [2]

    Bernton, J

    E. Bernton, J. Heng, A. Doucet, and P. E. Jacob. Schrödinger bridge samplers. arXiv preprint arXiv:1912.13170, 2019

    arXiv 1912

  3. [3]

    Blattmann, T

    A. Blattmann, T. Dockhorn, S. Kulal, D. Mendelevitch, M. Kilian, D. Lorenz, Y . Levi, Z. English, V . V oleti, A. Letts, et al. Stable video diffusion: Scaling latent video diffusion models to large datasets. arXiv preprint arXiv:2311.15127, 2023

    arXiv 2023

  4. [4]

    K. F. Caluya and A. Halder. Wasserstein proximal algorithms for the schrödinger bridge problem: Density control with nonlinear drift. IEEE Transactions on Automatic Control, 67(3):1163–1178, 2021

    work page 2021

  5. [5]

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

    arXiv 2023

  6. [6]

    Chen, G.-H

    T. Chen, G.-H. Liu, and E. A. Theodorou. Likelihood training of schr\" odinger bridge using forward-backward sdes theory.International Con- ference on Learning Representations, 2022

    work page 2022

  7. [7]

    Y . Chen, T. Georgiou, and M. Pavon. Entropic and displacement in- terpolation: a computational approach using the hilbert metric. SIAM Journal on Applied Mathematics, 76(6):2375–2396, 2016

    work page 2016

  8. [8]

    Z. Chen, G. He, K. Zheng, X. Tan, and J. Zhu. Schrodinger bridges beat diffusion models on text-to-speech synthesis. arXiv preprint arXiv:2312.03491, 2023

    arXiv 2023

Show all 50 references
  1. [9]

    Y . Choi, Y . Uh, J. Yoo, and J.-W. Ha. Stargan v2: Diverse image synthe- sis for multiple domains. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 8188–8197, 2020

    2020

  2. [10]

    De Bortoli, J

    V . De Bortoli, J. Thornton, J. Heng, and A. Doucet. Diffusion schrödinger bridge with applications to score-based generative model- ing. Advances in Neural Information Processing Systems , 34:17695– 17709, 2021

    2021

  3. [11]

    Dhariwal and A

    P. Dhariwal and A. Nichol. Diffusion models beat gans on image syn- thesis. Advances in neural information processing systems , 34:8780– 8794, 2021

    2021

  4. [12]

    H. Föllmer. Random fields and diffusion processes. In École d’Été de Probabilités de Saint-Flour XV–XVII, 1985–87 , pages 101–203. Springer, 1988

    1985

  5. [13]

    R. Fortet. Résolution d’un système d’équations de M. Schrödinger. Journal de Mathématiques Pures et Appliqués , 1:83–105, 1940

    1940

  6. [14]

    S. Gu, D. Chen, J. Bao, F. Wen, B. Zhang, D. Chen, L. Yuan, and B. Guo. Vector quantized diffusion model for text-to-image synthe- sis. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 10696–10706, 2022

    2022

  7. [15]

    Gushchin, D

    N. Gushchin, D. Selikhanovych, S. Kholkin, E. Burnaev, and A. Ko- rotin. Adversarial schrödinger bridge matching. Advances in Neural Information Processing Systems, 37:89612–89651, 2024

    2024

  8. [16]

    D. He, Y . Xia, T. Qin, L. Wang, N. Yu, T.-Y . Liu, and W.-Y . Ma. Dual learning for machine translation. Advances in neural information pro- cessing systems, 29, 2016

    2016

  9. [17]

    Heusel, H

    M. Heusel, H. Ramsauer, T. Unterthiner, B. Nessler, and S. Hochreiter. Gans trained by a two time-scale update rule converge to a local nash equilibrium. Advances in neural information processing systems , 2017

    2017

  10. [18]

    J. Ho, A. Jain, and P. Abbeel. Denoising diffusion probabilistic mod- els. Advances in neural information processing systems, 33:6840–6851, 2020

    2020

  11. [19]

    J. Ho, W. Chan, C. Saharia, J. Whang, R. Gao, A. Gritsenko, D. P. Kingma, B. Poole, M. Norouzi, D. J. Fleet, et al. Imagen video: High definition video generation with diffusion models. arXiv preprint arXiv:2210.02303, 2022

    2022 arXiv

  12. [20]

    Isola, J.-Y

    P. Isola, J.-Y . Zhu, T. Zhou, and A. A. Efros. Image-to-image translation with conditional adversarial networks. In Proceedings of the IEEE con- ference on computer vision and pattern recognition , pages 1125–1134, 2017

    2017

  13. [21]

    Jun and A

    H. Jun and A. Nichol. Shap-e: Generating conditional 3d implicit func- tions. arXiv preprint arXiv:2305.02463, 2023

    2023 arXiv

  14. [22]

    Karras, M

    T. Karras, M. Aittala, T. Aila, and S. Laine. Elucidating the design space of diffusion-based generative models. Advances in Neural Information Processing Systems, 35:26565–26577, 2022

    2022

  15. [23]

    J. Kim, M. Kim, H. Kang, and K. H. Lee. U-gat-it: Unsupervised gen- erative attentional networks with adaptive layer-instance normalization for image-to-image translation. In International Conference on Learn- ing Representations, 2020

    2020

  16. [24]

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

    2014 arXiv

  17. [25]

    Korotin, N

    A. Korotin, N. Gushchin, and E. Burnaev. Light schrödinger bridge. International Conference on Learning Representations , 2024

    2024

  18. [26]

    Kullback

    S. Kullback. Probability densities with given marginals. The Annals of Mathematical Statistics, 39(4):1236–1243, 1968

    1968

  19. [27]

    B. Li, K. Xue, B. Liu, and Y .-K. Lai. Bbdm: Image-to-image trans- lation with brownian bridge diffusion models. In Proceedings of the IEEE/CVF conference on computer vision and pattern Recognition , pages 1952–1961, 2023

    1952

  20. [28]

    S. Lin, B. Liu, J. Li, and X. Yang. Common diffusion noise schedules and sample steps are flawed. In Proceedings of the IEEE/CVF Win- ter Conference on Applications of Computer Vision , pages 5404–5411, 2024

    2024

  21. [29]

    Lipman, R

    Y . Lipman, R. T. Chen, H. Ben-Hamu, M. Nickel, and M. Le. Flow matching for generative modeling. International Conference on Learn- ing Representations, 2023

    2023

  22. [30]

    G.-H. Liu, A. Vahdat, D.-A. Huang, E. A. Theodorou, W. Nie, and A. Anandkumar. I 2 sb: Image-to-image schrödinger bridge. Inter- national Conference on Machine Learning , 2023

    2023

  23. [31]

    M.-Y . Liu, X. Huang, A. Mallya, T. Karras, T. Aila, J. Lehtinen, and J. Kautz. Few-shot unsupervised image-to-image translation. In Pro- ceedings of the IEEE/CVF international conference on computer vision, pages 10551–10560, 2019

    2019

  24. [32]

    X. Liu, C. Gong, and Q. Liu. Flow straight and fast: Learning to gen- erate and transfer data with rectified flow. International Conference on Learning Representations, 2023

    2023

  25. [33]

    Z. Liu, P. Luo, X. Wang, and X. Tang. Deep learning face attributes in the wild. In Proceedings of International Conference on Computer Vision (ICCV), December 2015

    2015

  26. [34]

    C. Lu, Y . Zhou, F. Bao, J. Chen, C. Li, and J. 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

  27. [35]

    Video generation models as world simulators

    OpenAI. Video generation models as world simulators. https: //openai.com/research/video-generation-models-as-world-simulators. Accessed: 2024-03-01

    2024

  28. [36]

    Pavon, G

    M. Pavon, G. Trigila, and E. G. Tabak. The data-driven schrödinger bridge. Communications on Pure and Applied Mathematics , 74(7): 1545–1573, 2021

    2021

  29. [37]

    Peluchetti

    S. Peluchetti. Bm2: Coupled schrödinger bridge matching. arXiv preprint arXiv:2409.09376, 2024

    2024 arXiv

  30. [38]

    Rombach, A

    R. Rombach, A. Blattmann, D. Lorenz, P. Esser, and B. Ommer. High- resolution image synthesis with latent diffusion models. InProceedings of the IEEE/CVF conference on computer vision and pattern recogni- tion, pages 10684–10695, 2022

    2022

  31. [39]

    Russakovsky, J

    O. Russakovsky, J. Deng, H. Su, J. Krause, S. Satheesh, S. Ma, Z. Huang, A. Karpathy, A. Khosla, M. Bernstein, et al. Imagenet large scale visual recognition challenge. International journal of computer vision, 115:211–252, 2015

    2015

  32. [40]

    Schrödinger

    E. Schrödinger. Sur la théorie relativiste de l’électron et l’interprétation de la mécanique quantique. Annales de l’Institut Henri Poincaré , 2(4): 269–310, 1932

    1932

  33. [41]

    Y . Shi, V . De Bortoli, A. Campbell, and A. Doucet. Diffusion schrödinger bridge matching. Advances in Neural Information Process- ing Systems, 36, 2024

    2024

  34. [42]

    Song and S

    Y . Song and S. Ermon. Generative modeling by estimating gradients of the data distribution. Advances in neural information processing sys- tems, 32, 2019

    2019

  35. [43]

    Taigman, A

    Y . Taigman, A. Polyak, and L. Wolf. Unsupervised cross-domain im- age generation. International Conference on Learning Representations, 2017

    2017

  36. [44]

    Z. Tang, S. Gu, C. Wang, T. Zhang, J. Bao, D. Chen, and B. Guo. V ol- umediffusion: Flexible text-to-3d generation with efficient volumetric encoder. arXiv preprint arXiv:2312.11459, 2023

    2023 arXiv

  37. [45]

    Vargas, P

    F. Vargas, P. Thodoroff, A. Lamacraft, and N. Lawrence. Solving schrödinger bridges via maximum likelihood. Entropy, 23(9):1134, 2021

    2021

  38. [46]

    P. Vincent. A connection between score matching and denoising au- toencoders. Neural Computation, 23(7):1661–1674, 2011

    2011

  39. [47]

    C. Wah, S. Branson, P. Welinder, P. Perona, and S. Belongie. The caltech-ucsd birds-200-2011 dataset. 2011

    2011

  40. [48]

    Z. Yi, H. Zhang, P. Tan, and M. Gong. Dualgan: Unsupervised dual learning for image-to-image translation. In Proceedings of the IEEE international conference on computer vision , pages 2849–2857, 2017

    2017

  41. [49]

    L. Zhou, A. Lou, S. Khanna, and S. Ermon. Denoising diffusion bridge models. International Conference on Learning Representations , 2024

    2024

  42. [50]

    J.-Y . Zhu, T. Park, P. Isola, and A. A. Efros. Unpaired image-to-image translation using cycle-consistent adversarial networks. In Proceedings of the IEEE international conference on computer vision , pages 2223– 2232, 2017

    2017

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.