- Rating
- 3
- Confidence
- 1
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
The paper studies when to apply reward guidance during flow-model sampling. It measures how the model responds to changes along the reward direction, then applies guidance during the period it identifies as most useful. For hard-constraint PDE tasks, it also projects the sample onto the feasible set later in sampling. The experiments cover PDE sampling and data assimilation. They show improvements on several tested benchmarks.
Critical strengths
1、The paper asks a useful question: when should guidance be applied ?
2、The directional measure offers a different view from measuring overall expansion or contraction.
3、The method combines guidance with late-stage projection for hard physical constraints.
4、Results on the tested scientific tasks are promising, including a longer Navier–Stokes rollout
Critical weaknesses
1、 All experiments are on scientific flow models. The paper shows physical fields, but does not test image or video generation models.
2、The theory uses a simple two-mode Gaussian model. The paper does not show that the same “mode commitment” explanation applies to its PDE and data-assimilation tasks.
3、The effect of the guidance window is hard to isolate. The PDE method also changes when projection is applied. Compared with PCFM, this means the results do not show which change drives the gains.
4、The data-assimilation test set is small. Each task uses eight trajectories. The SEVIR gains over FlowDAS are small relative to the reported variation.
5、The method uses adjoint optimization and repeatedly estimates its guidance window. The paper does not report the resulting runtime or compute cost.
Questions for authors
-
Is there an existing comparison that changes only the guidance window while keeping the rest of the method fixed?
-
Can the authors report the runtime or computational cost of the proposed method and main baselines?
-
On how many of the eight test trajectories does the proposed method outperform FlowDAS?
2 comments
Rebuttal for Reviewer B9iw [2/2]
Q2
We reports the per-sample wall-clock time for all methods. Our method shows a large runtime variance across samples which is caused by the trajectory-adaptive window. The length of the controlled segment, and hence the optimization cost, differs across samples.
Table 1. Per-sample wall-clock time in seconds.
| Family | Method | Heat | Burgers | RD1D | Optimization budget (iterations × steps) |
|---|---|---|---|---|---|
| Optimization-based | D-Flow | 11.79 ± 0.37 | 11.98 ± 0.39 | 12.35 ± 0.37 | 1000 |
| OC-Flow | 14.18 ± 0.30 | 14.70 ± 0.32 | 14.50 ± 0.32 | 1000 | |
| Projection-based | ECI | 8.41 ± 0.04 | 21.97 ± 0.16 | 26.47 ± 0.06 | – |
| PCFM | 12.07 ± 0.25 | 47.11 ± 0.58 | 26.56 ± 0.36 | – | |
| Both | Ours | 7.91 ± 1.02 | 55.64 ± 8.74 | 37.95 ± 2.36 | 150 / 600 / 600 |
We first compare the execution time. Since our method combines adjoint optimization with projection, we compare it against both families of baselines. Our method is the fastest on Heat. On Burgers and RD1D, its runtime is comparable to that of the projection-based baselines: it takes 1.18× and 1.43× the time of PCFM, on which our PDE pipeline is built. It is, however, slower than the optimization-based baselines D-Flow and OC-Flow on these two tasks.
Importantly, the performance gains of our method are not obtained by spending more optimization. When we compare the optimization budget for the adjoint updates (iterations × ODE steps), D-Flow and OC-Flow use 1000 step-updates on every task, whereas ours uses only 150 / 600 / 600 on Heat / Burgers / RD1D. Despite this smaller budget, our method outperforms both optimization-based baselines. This indicates that the gains come from when the adjoint-based control is applied rather than from how much optimization is performed. In addition, our method intervenes on a shorter portion of the trajectory, since the control is applied only inside the window [t̂_peak, t̂_c), while D-Flow and OC-Flow optimize over the entire trajectory.
We will include this table and discussion in the revised manuscript.
Q3
We thank the reviewer for this question. According to the comment, we count the number of improved trajectory by our method compared FlowDAS. As shown in the table, all 8 trajectories on both NSE tasks and SEVIR are improved by our method. This shows that the improvement shared by nearly all trajectories, not from a few outliers.
| NSE SR 8× | NSE SO 5% | SEVIR SO 10% | |
|---|---|---|---|
| Improved trajectory | 8/8 | 8/8 | 8/8 |
Rebuttal for Reviewer B9iw [1/2]
Q1.
Thanks for the thoughtful comment. To isolate the contribution of the window, we present the result for an ablation study.
[Data Assimilation] We first change the window’s boundaries, and move the window to the start point(t=0), the end point(t=1), or a random position. "Controlled steps" counts the guided steps (out of 20) per assimilation step, averaged over the 10 rollout steps. We report RMSE for first k∈{3,6,10} steps, to present errors over longer horizons.
We summarize the key observations that support our method as follows.
[PDE] We conduct the ablation on the Burgers equation benchmark. The comparison has three parts: (i) removing one component at a time (adjoint guidance or projection after tc). (ii) changing the window boundaries [te,tc), and (iii) full-trajectory control [0,1),
We summarize key observations as follows.