- Rating
- 2
- Confidence
- 3
- Informativeness
- 1
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper proposes CacheFlow, a frequency-aware framework for extreme few-step diffusion acceleration. The authors attribute quality degradation to asynchronous frequency evolution and introduce two main components: frequency-aware policy distillation for adaptive multi-band trajectory supervision, and a lightweight learnable cache policy for modeling cross-step feature dynamics. The two components are jointly optimized to reduce both distillation and caching errors. Experiments on several recent text-to-image and image-editing models demonstrate competitive visual quality with up to 55.31× speedup.
Critical strengths
-
The proposed framework provides a unified training scheme that integrates policy distillation with learnable feature caching, which is conceptually simple and easy to understand.
-
The paper conducts extensive experiments on several modern image generation models, including FLUX.1-dev, Qwen-Image, Qwen-Image-2.1, and Qwen-Image-Edit.
Critical weaknesses
-
The paper points out that the denoising process exhibits a distinction between low- and high-frequency information. However, this phenomenon has already been extensively studied in early DDPM and U-Net-based models. Therefore, it should not be considered a core contribution of this work.
-
I have some concerns regarding the results presented in Figure 3. First, the paper does not specify how much data was used to obtain these statistics, making it difficult to assess the reliability and generality of the reported results. Moreover, the paper does not provide a clear definition or detailed explanation of the so-called “Normalized Change” metric. Without a precise formulation or justification, this metric appears to be an ad hoc measure introduced by the authors.
-
As shown in Figure 3, the low-frequency component first decreases and then increases throughout the denoising process. Therefore, the results do not necessarily support the claim that the later stages of denoising primarily focus on high-frequency details. The observed trend seems to suggest a more complex evolution of frequency components during denoising, which warrants further analysis.
-
The interval approximation method adopted in this paper is somewhat similar to that used in MeanFlow. However, the authors do not appear to provide a direct quantitative comparison with MeanFlow. It would be helpful to include such a comparison to better demonstrate the advantages and differences of the proposed method.
-
Overall, the proposed method appears to be largely an A+B combination of step distillation and feature caching, with relatively limited methodological novelty.
Questions for authors
Please see the Critical Weaknesses.
6 comments
Response to Reviewer ubjq Part 2/4
W2/W3. Additional details and interpretation of Figure 3
A2 & A3: Thank you for pointing this out. We apologize for omitting the details of Figure 3 from the main text due to space constraints. Figure 3 examines how the variation of teacher velocity differs across frequency bands and denoising stages. This observation motivates the band-specific velocity policy in CacheFlow, which separately models the temporal evolution of low-frequency structure and high-frequency details, acting as an implicit, dynamic multi-scale reweighting mechanism. We have added these details and analyses in Lines 209–234 and 250–259 in our revised paper.
1. Metric definition and statistics for Figure 3
Normalized Change measures the relative magnitude of velocity changes within each frequency band between adjacent denoising steps. For the velocity along a teacher denoising trajectory, Haar DWT yields the low- and high-frequency components . We normalize the magnitude of the change between adjacent steps to obtain , and average the scores over the denoising trajectories. A higher Normalized Change indicates a larger normalized cross-step velocity change, whereas a lower value indicates greater stability between adjacent steps.
The statistics cover all prompts from GenEval, DPG-Bench, and OneIG-Bench, totaling 3,972 prompts. These prompts span diverse generation requirements, including compositional relationships, long textual descriptions, text rendering, styles, and reasoning.
2. Denoising progresses from low-frequency structure to high-frequency details
Figure 3 shows the changes of low- and high-frequency components at different denoising stages. Early in denoising, the normalized change of low-frequency components is substantially higher, reflecting the rapid evolution of global layout, contours, and smooth structures. As denoising progresses, the normalized change of high-frequency components gradually exceeds and becomes particularly pronounced in the later stages, reflecting the rapid evolution of edges, textures, and local details. Although the low-frequency components exhibit a late-stage increase, their normalized change remains clearly below that of high-frequency components. Thus, high-frequency components continue to dominate late-stage relative changes, reflecting a transition from establishing global structure early to refining details later in generation.
3. Figure 3 motivates our frequency policy as an implicit dynamic reweighting mechanism
Figure 3 shows that temporal dynamics are highly non-uniform across different noise levels. However, existing monolithic spatial parameterization mixes frequency bands within the same output channels, making it prone to spectral bias and capacity competition. Without carefully designed handcrafted timestep weighting of the loss, the model struggles to accommodate the evolving dynamics across different denoising stages.
Building on this observation, we propose a frequency-aware velocity policy that acts as an implicit, dynamic multi-scale reweighting mechanism that better accommodates the optimization demands of few-step distillation. Specifically, by decomposing the velocity into distinct Haar-wavelet bands and predicting band-specific parameters , our policy structurally decouples the gradient paths across frequency bands in the network head. As shown in Figure 3(c)–(d) in our revised paper, our frequency policy supervises the establishment of global structure with higher low-frequency loss weights during early denoising, and the refinement of local details with higher high-frequency loss weights during late denoising, consistent with our observations in Figure 3(a)–(b). CacheFlow adaptively learns the evolution rates at different scales and provides implicit dynamic reweighting during optimization, better accommodating the evolving dynamics across different denoising stages. Tables 8 and 5 in our revised paper further demonstrate that CacheFlow achieves the best velocity fitting accuracy and generation quality.
Due to space constraints, we continue in Response Part 3.
Response to Reviewer ubjq Part 3/4
W4. Differences from and comparison with MeanFlow
A4: MeanFlow reduces sampling steps through interval-average velocity prediction, but it faces training instability, additional computational overhead, and challenges in integrating feature caching. CacheFlow instead performs few-step policy distillation by predicting and directly matching the teacher's instantaneous velocity trajectory and integrates learnable feature caching under a unified objective, and its performance has been validated on multiple large-scale generative backbones. Details are provided below. We have added the comparison to Table 2, Lines 282–292 in our revised paper, with discussion in Lines 415–416.
1. Differences between CacheFlow and MeanFlow
MeanFlow represents interval displacement through the average velocity and constructs its training target using . However, the additional JVP term amplifies noise in the stochastic conditional velocity, causing training instability and increase in training overhead. Besides, its average-velocity prediction makes it difficult to precisely supervise the cache predictor at individual time points. CacheFlow instead avoids the JVP term, constructs a bandwise velocity policy , predicts and directly matches the teacher's instantaneous velocity at multiple intermediate time points through , and advances the trajectory through low-cost substeps. CacheFlow further jointly supervises the student and cache predictor at multiple intermediate times, enabling co-adaptation that reduces caching errors while preserving generation quality.
2. CacheFlow outperforms MeanFlow across multiple settings
We further compare CacheFlow and MeanFlow on FLUX.1-dev. As shown in Table 1, at 4 NFE, MeanFlow incurs drops of 3.0%/1.6% on GenEval/DPG-Bench relative to the original model, preserving its generation quality. However, when NFE is further reduced to 2 and 1, the two metrics drop substantially by 28.8%/11.4% and 53.0%/20.3%, indicating that MeanFlow struggles to maintain generation quality in large-scale diffusion models with very few sampling steps. In contrast, CacheFlow achieves 0.66/84.72 with only 2 NFE, delivering a 22.89× speedup. With caching, CacheFlow-Medium and CacheFlow-Fast further reduce the computational budget to 1.75/1.58 NFE and increase the speedup to 26.34×/29.20×, while maintaining scores of 0.66/84.36 and 0.65/84.08, demonstrating CacheFlow's substantial advantages in generation with very few steps.
Table 1: Comparisons with MeanFlow.
| Method | NFE ↓ | Latency ↓ | Speed ↑ | GenEval ↑ | DPG-Bench ↑ |
|---|---|---|---|---|---|
| FLUX.1-dev | 50 | 9.815 | 1.00× | 0.66 | 84.16 |
| MeanFlow | 4 | 0.853 | 11.50× | 0.64 | 82.79 |
| MeanFlow | 2 | 0.427 | 22.99× | 0.47 | 74.57 |
| MeanFlow | 1 | 0.214 | 45.97× | 0.31 | 67.04 |
| CacheFlow (w/o Cache) | 2 | 0.429 | 22.89× | 0.66 | 84.72 |
| CacheFlow-Medium | 1.75 | 0.373 | 26.34× | 0.66 | 84.36 |
| CacheFlow-Fast | 1.58 | 0.336 | 29.20× | 0.65 | 84.08 |
Due to space constraints, we continue in Response Part 4.
Response to Reviewer ubjq Part 1/4
Thank you very much for your detailed review and valuable feedback. We appreciate your suggestions for improving the clarity and quality of our work.
W1. Research scope and contributions of CacheFlow
A1: Thank you for your valuable suggestion. We acknowledge prior studies of frequency evolution in diffusion models. However, our work investigates the joint optimization of step distillation and learnable feature caching in modern Flow Matching models with DiT backbones. The co-adaptation of the distilled student and cache predictor under sparse sampling remains beyond the scope of these earlier studies.
Earlier U-Net architectures extract multiscale features through convolution, downsampling, and skip connections, whereas DiTs aggregate spatial information through attention softmax(QK⊤/d)V. DDPMs learn noise predictions along a stochastic forward diffusion process, whereas the Flow Matching models learn velocity fields along interpolation paths xt=(1−t)x0+tϵ. These differences affect both feature representations and denoising dynamics.
Motivated by these differences, our work investigates the joint modeling of trajectory distillation and feature caching in modern diffusion models based on Flow Matching and DiT. Our observations of distinct velocity dynamics across frequency bands motivate a frequency-aware policy, and we further jointly optimize frequency-aware policy distillation and learnable feature caching under unified supervision, improving teacher-velocity matching and reducing caching errors in few-step generation.
Due to space constraints, we continue in Response Part 2.