- Rating
- 2
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
The paper post-trains the exact few-step sampling map of a flow-matching model for a given solver and NFE. In Stage I, a terminal distribution loss is differentiated through the deployed solver; the loss is a Fréchet distance on projected Inception pool-3 features, plus FM anchor terms. This updates a late subset of the network and a 16-logit monotone time reparameterization ("clock") shared across Heun, RK3, and RK4. In Stage II, the network is frozen, small solver-specific residual clocks are learned, and a held-out alignment statistic decides whether each one is deployed. On a C2OT CIFAR-10 model at NFE=12, mean CleanFID10k drops from 46.8 to 7.70 after Stage I and to 7.33 after Stage II. On SiT-XL/2, a finite-stage temporal operator improves Heun and RK4 FID50k by 0.56 and 0.14. Further experiments cover B-series analysis, transfer across budgets and RK2 tableaus, composition with STORK, and four non-image domains.
Critical strengths
- Optimizing the deployed solver- and budget-specific map is well motivated. The method adds no NFE at deployment and keeps a single field across solvers.
- The B-series analysis and the NFE-decay study (5.586% → 2.513% → 0.250%) support the claim that the gains come from finite-step error.
- Statistical practice is careful (paired streams, multiple seeds, rules frozen before confirmation, 50k-sample FID), and the coverage is broad, including non-image domains.
Critical weaknesses
- The Stage I objective is a Fréchet distance on CleanFID Inception features, essentially the same network as the evaluation metric. The headline C2OT gain (46.8 → 7.7) may therefore partly reflect metric fitting.
- Quality is not competitive: BézierFlow reaches 6.19/5.08/3.96 FID50k at NFE=8/12/24, versus 20.56/9.43/5.49 for the proposed method (Sec. 5.9). Optimized schedules and bespoke solvers are not compared.
- Stage II adds considerable machinery for small gains (4.8% on C2OT, none for Heun), and heavy non-standard terminology makes the paper hard to follow.
Questions for authors
- Can you report the main C2OT and SiT results with features independent of the training objective (e.g., DINOv2 FD and CLIP-FID for all rows)?
- How does the method compare with an optimized timestep schedule (Sabour et al., 2024) or bespoke solvers (Shaul et al., 2024) at the same NFE and a similar adaptation cost?
- On SiT, how much benefit is lost if Stage II is removed?
5 comments
B. Further comparison with BézierFlow
We directly compared against the BézierFlow SiT Compatibility Port at NFE12.
| Method | ADM-FID50k ↓ | DINOv2 FD ↓ | CLIP-FID ↓ |
|---|---|---|---|
| BézierFlow SiT Compatibility Port | 3.184069 | 182.942166 | 3.379921 |
| Ours: Shared Adaptation (RK4) | 2.853464 | 168.864649 | 3.274840 |
Our RK4 configuration achieves lower values on all three distances in this comparison.
We also checked different BézierFlow training configurations.
| BézierFlow Configuration | FID |
|---|---|
| Same trajectory, 125 updates | 3.2237 |
| Same trajectory, 500 updates | 3.2243 |
| Increased to 200 training pairs | 3.1513 |
| Ours: Shared Adaptation (RK4) | 2.8535 |
| Ours: Shared Model + Midpoint | 2.8420 |
Increasing BézierFlow updates from 125 to 500 on the same trajectory did not improve FID. Increasing the number of training pairs to 200 reduced FID, but it remained above Shared RK4.
Deploying the same shared model with Midpoint gives FID 2.8420.
C. Optimized schedules and specialized solvers
We added ImageNet-256 comparisons with training-free samplers, optimized schedules, and learned specialized solvers at NFE12.
| Method | ADM-FID50k ↓ | DINOv2 FD ↓ | CLIP-FID ↓ |
|---|---|---|---|
| STORK Order-2 | 3.131723 | 182.870190 | 3.293903 |
| STORK + Frozen Clock | 3.087859 | 181.441707 | 3.272954 |
| Bespoke RK2 Full-port | 2.884906 | 179.971207 | 3.155823 |
| Bespoke RK2 Resource-cut | 2.911586 | 181.370197 | 3.167371 |
| Ours: Shared Adaptation (Heun) | 3.993210 | 197.755778 | 3.990408 |
| Ours: Shared Adaptation (RK4) | 2.853464 | 168.864649 | 3.274840 |
Shared RK4 has lower ADM-FID50k and DINOv2 FD than both Bespoke configurations, while Bespoke has better CLIP-FID.
Shared RK4 also has lower ADM-FID50k and DINOv2 FD than STORK, with slightly higher CLIP-FID.
The supplement reports full results for Uniform Native Midpoint and the AYS MC128 Resource Port.
These results support competitiveness in the tested low-NFE ImageNet configurations.
D. Quality at similar adaptation cost
To address your request, we measured acquisition costs on ImageNet-256.
| Method | Acquisition Cost (seconds) |
|---|---|
| Uniform Native Midpoint | 0 |
| AYS-flow-grid Heun (MC128 Port) | 7,463.81 |
| Bespoke RK2 Full-budget Port | 15,034.45 |
| Bespoke RK2 Resource-cut Port | 9,571.74 |
| Ours: Shared Adaptation (Stage I only) | 7,822.59 |
| Ours: Full Pipeline (Stage I + Residual Candidates + Selection) | 9,574.85 |
The full procedure costs 9,574.85 seconds, only 3.11 seconds more than Bespoke Resource-cut's 9,571.74 seconds. At this similar cost, Shared RK4 has lower ADM-FID50k and DINOv2 FD, while Bespoke has better CLIP-FID.
Stage I alone costs 7,822.59 seconds. Stage II adds 1,752.26 seconds for candidate acquisition and selection.
In this SiT setting, the prespecified development selection rule chooses the exact Stage-I fallback for both Heun and RK4. Removing Stage II preserves final deployment quality and saves its acquisition cost.
E. Interaction between training and deployment solvers
We added a full 2×2 experiment on CIFAR-10, using Heun and Midpoint as both training and deployment solvers.
The table summarizes FID50k for all four combinations.
| Training solver | Heun deployment | Midpoint deployment |
|---|---|---|
| Heun | 8.9048 ± 0.1904 | 6.4433 ± 0.2545 |
| Midpoint | 36.4151 ± 1.1783 | 4.0793 ± 0.0562 |
The benefit of a training solver depends on the deployment solver. Heun training gives lower FID with Heun deployment, while Midpoint training gives lower FID with Midpoint deployment.
The original CIFAR-10 Heun result therefore does not establish the framework's performance limit. The controlled Midpoint experiment achieves better quality, and the full cross experiment supports a substantial interaction between training and deployment solvers.
We will report FID50k, DINOv2 FD, and CLIP-FID for all four combinations in the supplement and account for the interaction when interpreting solver performance.
F. Method scope and contributions
BézierFlow focuses on absolute quality for few-step generation. Our work addresses adapting pretrained Flow Matching models to practical low-NFE deployment.
Our method has three main features.
(a) Shared field-and-clock adaptation
We jointly adjust part of the velocity field and a shared time parameterization to adapt finite-step generation without an external teacher.
(b) Reuse across configurations and budgets
On CIFAR-10, one shared adaptation costs 281.55 seconds and is reused across eight prespecified solver/budget configurations.
On ImageNet-256, the frozen finite-stage temporal operator acquired at NFE12 transfers directly to NFE8 and NFE24 without refitting.
(c) Conditional solver-specific refinement
Residual refinement is retained only when validation supports it. Otherwise, deployment returns exactly to the shared adaptation.
Our contribution is a deployment-aware post-training framework for pretrained Flow Matching models. Shared adaptation, reuse across configurations, and optional specialization improve low-step deployment. We do not claim superiority to BézierFlow across all datasets and NFE settings.
Comment 3. Stage-II complexity and benefit
Question. Stage II adds complexity for a small gain. How much performance is lost if Stage II is removed on SiT?
Response 3
Thank you for this question.
A. Direct comparison on SiT-XL/2 ImageNet-256
| Solver | Stage-I FID | Raw Residual FID | Final selection |
|---|---|---|---|
| Heun | 3.993210 | 4.000501 | Stage-I Fallback |
| RK4 | 2.853464 | 2.855850 | Stage-I Fallback |
Neither raw residual candidate improves on Stage I. The development selection rule, fixed before formal evaluation, chooses the exact Stage-I fallback for both solvers.
Thus, removing Stage II causes no deployment-quality loss in this SiT setting.
Removing residual training and selection reduces the shared acquisition cost from 9,574.851 to 7,822.585 seconds, saving 1,752.266 seconds. This is a one-time shared acquisition and selection saving. Inference speed is unaffected, and the saving is not counted separately for each solver.
B. Larger, repeatable gains on C2OT–RK3
The table shows NFE12 FID50k results on the original paired generation stream.
| Stage-I model | Stage I | With fixed Stage II | Relative FID reduction |
|---|---|---|---|
| Original selection model | 6.4163 | 5.4275 | 15.41% |
| Additional model 1 | 6.3271 | 4.9968 | 21.03% |
| Additional model 2 | 5.9179 | 4.7085 | 20.44% |
We will clarify that Stage I is the core of the method, and Stage II is optional solver-specific refinement. Its value depends on the model, solver, and deployment budget. Extra residual parameters do not guarantee better quality.
Comment 4. Nonstandard terminology
Question. Too many nonstandard terms make the method hard to understand.
Response 4
Thank you for pointing this out. We will use consistent terminology across the Abstract, Introduction, Methods, tables, and figure captions, and reduce repeated descriptions of the framework while retaining the mathematical definitions and reproducibility details. Please see the updated PDF.
Comment 1. Metric fitting in Stage I
Response 1
Thank you for raising the possible dependence between the training objective and evaluation metric.
We added DINOv2-L/14 FD and CLIP ViT-B/32 Fréchet distance to the main C2OT and SiT experiments, using the same saved images as the Inception evaluation.
The C2OT NFE12 results are below.
All shared-adaptation arms improve DINOv2 FD and CLIP-FID over the original model. Both active residual arms also improve these independent distances over their shared starting points.
The main gains therefore extend beyond the Inception feature space used in training.
We also tested training-feature and projection-dimension sensitivity using Inception, DINOv2, and CLIP features with 16-, 32-, and 64-dimensional projections.
All nine training configurations outperform the matched zero-fit control in the two feature spaces excluded from their training objective.
We will add these independent-feature evaluations and clearly state the training features, evaluation features, and sample counts for FID10k and FID50k.
Comment 2. Generation quality and external baselines
Response 2
Thank you for identifying these gaps. We considered the results of BézierFlow, optimized schedules, and specialized solvers, then added controlled comparisons and mechanism analysis on CIFAR-10 and ImageNet-256.
The new experiments show that performance depends strongly on the training solver, deployment solver, and their interaction with the adapted velocity field. We address quality, solver interaction, and deployment applicability below. Code is available in the anonymous repository.
A. The original CIFAR-10 quality gap
We agree with your observation about the original Heun configuration. It did not achieve a better FID than BézierFlow at NFE8, NFE12, or NFE24.
We then tested Midpoint training and Midpoint deployment with the same initial checkpoint, NFE12, training objective, model capacity, and 64 optimization updates.
All three distances improve.
All three metrics improve in all nine paired conditions formed by three independent training seeds and three independent generation streams.
The corresponding Heun training and deployment results in the same CIFAR-10 diagnostic are below.
The original gap is concentrated in the Heun configuration. It does not establish an overall lack of competitiveness for the adaptation framework on CIFAR-10.