- Rating
- 3
- Confidence
- 3
- Informativeness
- 3
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
This paper proposes iSSFM, an improved training framework for strong stochastic flow maps. The key idea is to use a shifted Legendre polynomial representation of Brownian paths together with a Wong–Zakai approximation, which transforms the underlying SDE into a smooth random ODE and enables derivative-based training objectives such as Improved MeanFlow. The authors further develop a fine-tuning strategy that converts pretrained flow models into strong stochastic flow maps.The method is evaluated on several toy problems and ImageNet-256, demonstrating substantial improvements over existing semigroup-based training.
Critical strengths
-
Interesting methodological contribution. The combination of shifted Legendre polynomial representations and Wong–Zakai approximation provides an effective way to extend derivative-based training objectives to strong stochastic flow maps. In particular, the proposed formulation enables the use of higher-order Brownian path information during training.
-
Strong empirical results. The proposed method substantially improves few-step sampling performance over the semigroup-based baseline. On ImageNet-256, it reduces FID from 12.97 to 5.36 at 4 NFE. The experiments also include useful comparisons of Brownian representations and training objectives.
Critical weaknesses
-
Limited evaluation of strong approximation accuracy. While the paper demonstrates strong improvements in sampling quality, the ImageNet experiments evaluate pathwise consistency against the model's own 32-step solution rather than a high-accuracy numerical solution of the pretrained SDE. A direct comparison against the underlying SDE would provide stronger evidence of the learned strong stochastic flow map's accuracy.
-
Limited downstream applications. The Feynman–Kac corrector experiment provides an interesting application, but it is restricted to a low-dimensional Gaussian mixture. The practical advantages of strong stochastic flow maps over deterministic or weak stochastic alternatives remain insufficiently demonstrated, particularly for high-dimensional generative modeling tasks.
Questions for authors
-
Applicability to pretrained diffusion models. Can the proposed Wong–Zakai-based training method be directly applied to pretrained score-based diffusion models? If so, what is the motivation or practical advantage of constructing an SDE from a pretrained flow matching model rather than directly learning the stochastic flow map of a pretrained diffusion SDE?
-
Necessity of strong stochastic flow maps. Could the authors elaborate on the practical importance of preserving strong pathwise correspondence? Beyond the Feynman–Kac corrector example, are there important downstream applications that specifically require strong stochastic flow maps rather than weak stochastic flow maps or deterministic alternatives? In particular, what advantages does the strong property provide that cannot be achieved by matching the appropriate transition distributions?
6 comments
Response to 5V2D part 2
Weakness. Limited downstream applications. The Feynman–Kac corrector experiment provides an interesting application, but it is restricted to a low-dimensional Gaussian mixture. The practical advantages of strong stochastic flow maps over deterministic or weak stochastic alternatives remain insufficiently demonstrated, particularly for high-dimensional generative modeling tasks.
Question. Necessity of strong stochastic flow maps. Could the authors elaborate on the practical importance of preserving strong pathwise correspondence? Beyond the Feynman–Kac corrector example, are there important downstream applications that specifically require strong stochastic flow maps rather than weak stochastic flow maps or deterministic alternatives? In particular, what advantages does the strong property provide that cannot be achieved by matching the appropriate transition distributions?
Yes, we find that for image editing with SDEs this pathwise property is crucial.
We ran some additional experiments (new Section 4.3, Figure 5, Table 4; Appendix H.4) doing editing of 240 real ImageNet-256 images consisting of 30 source to target class pairs each with 8 images. Every method received the same NFE budget , counting both inversion from the real image and generation. We map a source image to a noisy state , change the class condition, and generate from . DMF inverts its ODE, whereas the SDE methods noise the source forward, record the Brownian increments that carry it back to under the source class [1], and replay those increments under the new class; this requires a map that follows a given Brownian path. Methods are compared at equal preservation to the source image measured in LPIPS [3] from the original image. We compare DMF, DMF (denoise–renoise), the teacher SDE following [1] (Cycle-SDE), SSFM, and iSSFM, as well as SSFM (weak) and iSSFM (weak).
Table R.2. Image editing with different models (Table 8 of the revised PDF).
| Method | NFE | CLIP @ LPIPS 0.12 | CLIP @ 0.20 | PickScore @ 0.12 | PickScore @ 0.20 | ImageReward @ 0.12 | ImageReward @ 0.20 |
|---|---|---|---|---|---|---|---|
| DMF | 16 | 20.62 | 21.76 | 19.45 | 19.60 | −1.25 | −1.08 |
| Teacher SDE (Cycle-SDE) | 16 | 21.01 | — | 19.40 | — | −1.22 | — |
| SSFM (weak) | 16 | — | 19.65 | — | 19.26 | — | −1.40 |
| iSSFM (weak) | 16 | — | 19.73 | — | 19.35 | — | −1.36 |
| SSFM | 16 | 20.55 | 21.67 | 19.40 | 19.53 | −1.24 | −1.11 |
| iSSFM (ours) | 16 | 21.01 | 22.15 | 19.45 | 19.59 | −1.18 | −1.04 |
| DMF | 32 | 20.72 | 21.89 | 19.41 | 19.55 | −1.23 | −1.07 |
| Teacher SDE (Cycle-SDE) | 32 | 20.97 | 21.88 | 19.43 | 19.44 | −1.21 | −1.16 |
| SSFM (weak) | 32 | — | 19.67 | — | 19.22 | — | −1.40 |
| iSSFM (weak) | 32 | — | 19.82 | — | 19.32 | — | −1.36 |
| SSFM | 32 | 20.72 | 22.00 | 19.42 | 19.58 | −1.22 | −1.07 |
| iSSFM (ours) | 32 | 21.05 | 22.34 | 19.47 | 19.64 | −1.17 | −0.99 |
| Teacher SDE (Cycle-SDE), many-step reference | 38–104 | 21.04 | 21.94 | 19.44 | 19.48 | −1.21 | −1.19 |
N.B., entries with — denote that the method never preserves the source this well; even its mildest edit is further away. Note, even the mildest edit of DMF (denoise–renoise) has LPIPS 0.25 at NFE 16 and 0.30 at NFE 32 (row omitted from table since it failed). The teacher SDE uses the same recorded Brownian increments as iSSFM with Euler–Maruyama steps of the frozen teacher. The many-step reference runs on the full 64-step rollout, so its NFE ranges from 38 to 104 depending on . (weak) denotes the same network run from the noised source on an independent Brownian path instead of the recorded one, i.e., using only its transition distributions, thus a weak map.
Figure 5 of the revised PDF shows qualitative edits at 32 NFE: iSSFM changes the object while keeping the pose, layout, and background of the source, DMF and DMF (denoise–renoise) redraw the scene, and iSSFM (weak) keeps the class but not the layout.
We summarize the main takeaways:
- iSSFM edits best at matched NFE, leading DMF and SSFM on every metric at NFE 32 (Appendix H.4, Table 9).
- DMF (denoise–renoise) cannot edit conservatively: it largely creates an entirely new image.
- SSFM roughly ties DMF; our improvements are what make pathwise editing beat DMF.
- The strong property is necessary: the weak maps lose 2–2.5 CLIP at equal preservation and cannot edit conservatively.
- The teacher SDE needs many more steps and still trails iSSFM.
Additionally, concurrent work to ours [2] has explored SSFMs in physical time to accelerate molecular dynamics by learning to take larger jumps in solving the Langevin SDE.
Response to 5V2D part 3
Concluding remarks
We thank the reviewer for their insight and helpful suggestions. We believe that the new experiments exploring both the strong property and its applications for high-dimensional generative modeling tasks have greatly improved the quality of this paper. We kindly ask that in light of these new experiments and clarifications the reviewer would consider them in their review. We are happy to engage in further discussion.
References
[1] Nie, S., Guo, H. A., Lu, C., Zhou, Y., Zheng, C., & Li, C. (2024). The Blessing of Randomness: SDE Beats ODE in General Diffusion-based Image Editing. ICLR 2024.
[2] McCallum, S., Rindtorff, N., Tong, A., & Foster, J. (2026). Langevin Flow Maps: Efficient Molecular Dynamics and Transition Path Sampling. arXiv preprint arXiv:2610.05998.
[3] Zhang, R., Isola, P., Efros, A. A., Shechtman, E., & Wang, O. (2018). The Unreasonable Effectiveness of Deep Features as a Perceptual Metric. CVPR 2018.
Response to 5V2D part 1
We thank reviewer 5V2D for their thoughtful review of our paper and interest in the downstream applications. We are happy the reviewer found the work to be an "[i]nteresting methodological contribution" and to have "[s]trong empirical results". Below we answer the questions and concerns raised by the reviewer.
At the reviewer's suggestion we compared these methods against an underlying 512-step Euler–Maruyama solve (revised Figure 4b; Appendix H.3, Table 7). We compare each one of our core changes: the exponential lift, closed-form martingale, drift matching loss, and Wong–Zakai iMF loss to show the impact on strong error. N.B., as our parameterizations use an exponential lift we compare those to the exponentially lifted SDE Yt=αtXt.
Table R.1. Strong mean squared error w.r.t. the teacher SDE (Table 7 of the revised PDF).
Our improved parameterization (exponential lift, closed-form martingale, drift matching) yields the lowest strong error, and it is below the teacher's own lifted Euler–Maruyama solver at matched NFE. iSSFM, which adds the Wong–Zakai iMF loss, gives the best fidelity among the stochastic flow maps and converges strongly.
Yes we can do both! We chose to write the paper using the flow matching formalism but everything holds mutatis mutandis for score matching models using utθ(x)=αtα˙tx+2νt2=stθ(x)∇logptθ(x). It is merely a choice of how the base model was trained, did it learn the velocity, score, noise prediction, or data prediction? We can then fine-tune from any of these formalisms. We clarified this in footnote 2 of the revised PDF.