- Rating
- 2
- Confidence
- 4
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper introduces S⁴-dyn, a method that combines frequency-ordered scanning—a technique derived from Mamba—with spatial modulation based on historical physical fields to predict the evolution of physical fields governed by dynamic partial differential equations (PDEs) in an autoregressive manner. The authors extend this approach to three dimensions using tri-plane decomposition and evaluate its performance across 15 benchmark tasks, covering single-step prediction, multi-step rolling prediction, and the specific contributions of each individual module.
Critical strengths
-
Each component of the architecture serves a distinct purpose. A frequency-domain Mamba module models cross-frequency dependencies, local convolutions handle local spatial structures, and a historical branch generates modulation parameters—which vary according to spatial location—to assist in predicting the next time step.
-
The experimental scope is extensive. The paper evaluates the model across 15 2D and 3D benchmark tasks, encompassing various dynamical systems such as acoustic wave propagation, reaction-diffusion, fluid dynamics, and magnetohydrodynamics.
-
Key results provide quantitative validation. Across 13 systems in "The Well" benchmark—comprising 39 distinct single-step and multi-step prediction tasks—the proposed method achieved the lowest error among all compared approaches in 28 of these tasks.
-
Ablation studies confirm the utility of specific design elements. For instance, in the "Active Matter" task, removing the historical spatial modulation caused the average prediction error over the first 20 steps to increase from 0.5466 to 0.8040; furthermore, comparisons of different scanning strategies also validate the effectiveness of the proposed design.
Critical weaknesses
The methodological novelty appears limited. FourierMamba has already combined Fourier-domain modeling with Mamba in a closely related framework. The present paper primarily introduces component-level modifications and extends the evaluation to 3D data. In my view, these contributions are insufficient to establish a substantial methodological advance. Moreover, the experiments compare only FourierMamba’s scanning strategy, rather than the complete model, leaving the added value of the proposed extensions insufficiently demonstrated.
Questions for authors
Please address the concerns raised in the Weaknesses above.
5 comments
Response to 6CGr (2/3): History-conditioned modulation and 3D spectral modeling
Q1: continued
2.Spatial Feature Modulation Branch. Our Spatial Feature Modulation Branch uses the Spectral Update Layer to evolve historical observations in latent spectral space and predict the next-step latent state. The Spatial-FiLM Modulation Head then uses this predicted state to generate layer-wise, spatially varying modulation fields for the Spectral-Spatial State-Space Branch. This differs from FourierMamba's spatial and frequency feature processing, which does not use a predicted future latent state to generate modulation.
As shown in Table R3-2, removing the entire branch (w/o SFM) increases rollout errors on both datasets. Retaining the frame-wise encoder and Spatial-FiLM Modulation Head, we also replace the Spectral Update Layer with three approximately parameter-matched alternatives: Spatial update + FiLM (channel-wise temporal projection and local convolutions in latent space without FFT/IFFT), ConvGRU + FiLM (3×3 convolutional gates across frames; Ballas et al., 2016), and Temporal Transformer + FiLM (single-layer, four-head temporal self-attention per location). S⁴-dyn achieves lower 1:20 and 21:60 rollout VRMSE than all three alternatives on both datasets. On Active Matter and TRL 2D, it reduces the 1:20 VRMSE by 73.3% and 14.2% over Spatial update + FiLM, and by 20.0% (vs. Temporal Transformer + FiLM) and 9.5% (vs. ConvGRU + FiLM) over the stronger temporal encoder for 1:20 rollout on each dataset.
Table R3-2. Ablations of spatial modulation and latent state updates. Metrics are VRMSE.
| Dataset | Variant | Params | Next-step ↓ | 1:20 ↓ | 21:60 ↓ |
|---|---|---|---|---|---|
| TRL 2D | S⁴-dyn | 6.55M | 0.1342 | 0.4115 | 0.9263 |
| w/o SFM | 6.01M | 0.1361 | 0.6081 | 1.0027 | |
| Spatial update + FiLM | 6.55M | 0.1426 | 0.4797 | 0.9469 | |
| ConvGRU + FiLM | 6.54M | 0.1325 | 0.4547 | 1.0324 | |
| Temporal Transformer + FiLM | 6.56M | 0.1407 | 0.4889 | 0.9896 | |
| Active Matter | S⁴-dyn | 6.59M | 0.0126 | 0.5466 | 1.3731 |
| w/o SFM | 6.04M | 0.0159 | 0.8040 | 1.7271 | |
| Spatial update + FiLM | 6.59M | 0.0977 | 2.0462 | 1.7898 | |
| ConvGRU + FiLM | 6.58M | 0.0257 | 1.5978 | 1.3919 | |
| Temporal Transformer + FiLM | 6.59M | 0.0194 | 0.6836 | 1.4748 |
3.Tri-Plane Factorized Spectral Module. We extend spectral state-space modeling to 3D physical fields through BDRA processing on three orthogonal spectral planes and Tri-Plane Volumetric Fusion. This reduces the sequence length per scanning direction from to , avoiding dense volumetric scanning. On 3D Rayleigh–Taylor and TGC, the complete S⁴-dyn model reduces the 1:20 rollout VRMSE by 84.6% and 62.3% compared with Flower (Table R3-3).
Table R3-3. Prediction performance on 3D physical systems. The metric is VRMSE.
| Dataset | Model | Next-step ↓ | 1:20 ↓ | 21:60 ↓ |
|---|---|---|---|---|
| MHD 64 | Flower | 0.1165 | 0.7580 | 1.2827 |
| S⁴-dyn | 0.0834 | 0.7425 | 1.2160 | |
| Rayleigh-Taylor | Flower | 0.0491 | 0.5862 | 1.4880 |
| S⁴-dyn | 0.0283 | 0.0901 | 0.7511 | |
| TGC | Flower | 0.1700 | 1.2636 | 3.7477 |
| S⁴-dyn | 0.0777 | 0.4758 | 1.0062 |
[1] Nicolas Ballas, Li Yao, Chris Pal, and Aaron Courville. Delving Deeper into Convolutional Networks for Learning Video Representations. ICLR, 2016.
Response to 6CGr (3/3): Full-model comparison with FourierMamba on TRL 2D and Active Matter
Q2: Full-model comparison with FourierMamba.
To address the concern regarding the lack of a full-model comparison, we reproduced the complete FourierMamba architecture based on its original paper, rather than comparing only its zigzag scanning strategy. Since the original implementation is not publicly available, our reproduction retains the Fourier Residual State-Space Blocks (FRSSBs), the Fourier Spatial Interaction SSM (FSI-SSM), the Fourier Channel Evolution SSM (FCE-SSM), and the original scanning strategies. We only reshaped the input from to and adjusted the output channels to to accommodate autoregressive dynamic PDE forecasting.
For the full-model comparison on TRL 2D and Active Matter, we trained the reproduced FourierMamba using the same data splits and training settings as S⁴-dyn in the original submission, including relative L2 loss, AdamW optimizer, OneCycleLR scheduler, batch size of 4, and 100 epochs on an NVIDIA A100 GPU. As shown in Table R3-4, S⁴-dyn achieves lower VRMSE than the reproduced full FourierMamba model across all three metrics on both datasets, reducing the next-step, 1:20, and 21:60 rollout errors by 44.2%, 43.6%, and 27.5% on TRL 2D (6.55M vs. 10.03M parameters), and by 96.7%, 50.6%, and 8.5% on Active Matter (6.59M vs. 10.05M parameters).
Table R3-4. Full-model comparison on TRL 2D and Active Matter. Metrics are VRMSE.
| Dataset | Model | Params | Next-step ↓ | 1:20 ↓ | 21:60 ↓ |
|---|---|---|---|---|---|
| TRL 2D | Flower | 17.33M | 0.1930 | 0.5491 | 1.0516 |
| FourierMamba | 10.03M | 0.2407 | 0.7302 | 1.2770 | |
| S⁴-dyn | 6.55M | 0.1342 | 0.4115 | 0.9263 | |
| Active Matter | Flower | 17.34M | 0.0249 | 1.3905 | 1.5251 |
| FourierMamba | 10.05M | 0.3781 | 1.1068 | 1.5003 | |
| S⁴-dyn | 6.59M | 0.0126 | 0.5466 | 1.3731 |
In addition to the full-model comparison in Table R3-4, the controlled scanning-path comparison in Table R3-1 (keeping the S⁴-dyn framework unchanged) shows that BDRA reduces the 1:20 rollout VRMSE compared with FourierMamba zigzag from 0.9678 to 0.5466 on Active Matter and from 0.4417 to 0.4115 on TRL 2D, with fewer parameters.
Response to 6CGr (1/3): Differences from FourierMamba in spectral-spatial modeling
We thank the reviewer for the thorough review. Below, we clarify how S⁴-dyn differs from FourierMamba in spectral scanning, history-conditioned modulation, and 3D spectral modeling. We also provide a full-model comparison with our FourierMamba reproduction on TRL 2D and Active Matter under identical training settings, together with additional controlled scanning and history-conditioning ablations on Active Matter and TRL 2D.
Q1: Methodological novelty beyond FourierMamba.
The novelty of S⁴-dyn lies in three components tailored to dynamic PDE forecasting: the Spectral-Spatial Operator, which combines BDRA scan based spectral state-space modeling with local convolution; the Spatial Feature Modulation Branch, which uses latent spectral evolution to generate history-conditioned Spatial-FiLM; and the Tri-Plane Factorized Spectral Module, which extends spectral state-space modeling to 3D physical fields.
1.Spectral-Spatial Operator. Whereas FourierMamba operates on amplitude and phase spectra, our Bidirectional Spectral State-Space Operator updates the real and imaginary components of the retained Fourier modes in ΛR while preserving the complementary modes. Moreover, unlike FourierMamba's progressive and bilateral zigzag scanning, our BDRA scan groups the retained modes into radial rings from low to high frequencies and traverses each ring in two opposite angular directions. For spatial processing, FourierMamba uses a Spatial Mamba branch, whereas S⁴-dyn combines its Bidirectional Spectral State-Space Operator with a Local Convolutional Operator for local spatial interactions.
In Table R3-1, the component ablations w/o C(ℓ) and w/o GΛR(ℓ) show that removing either operator increases both rollout errors on Active Matter. For the scanning-path ablations, compared with the parameter-matched 2-dir raster scan, BDRA reduces the 1:20 rollout VRMSE by 20.2% on Active Matter and 24.4% on TRL 2D. Compared with the parameter-matched Reverse BDRA, which reverses only the radial scanning order from low-to-high to high-to-low, BDRA further reduces this error by 6.5% on Active Matter and 5.5% on TRL 2D. BDRA also outperforms w/o π− (unidirectional angular scan), w/ radial reversal (which adds two high-to-low radial paths), 4-dir raster scan, and FourierMamba zigzag in both rollout intervals on both datasets. The full-model comparison with FourierMamba is presented in Q2.
Table R3-1. Component and scanning-path ablations on Active Matter and TRL 2D. Metrics are VRMSE.