- Rating
- 3
- Confidence
- 2
- Informativeness
- 3
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
The paper proposes Decomposed Inference (specifically Polyphase Inference) as a way to transfer a neural PDE solver pretrained at low resolution to a higher resolution by decomposing the fine-grid input into multiple coarse-grid components, applying the pretrained model independently to each, and recomposing the outputs, thereby avoiding both the aliasing errors of direct high-resolution inference and the information loss of compressed inference. The authors provide a theoretical characterization showing that direct FNO inference suffers from cross-resolution discretization mismatch, that compressed inference has an irreducible error floor, and that component-wise marginal invariance is the key condition for risk preservation under decomposition. Experiments across multiple PDEs (Burgers, Darcy flow, Reaction-Diffusion, Compressible Navier-Stokes), architectures (FNO, UNet, Transolver, LaMO), and large pretrained physics models (DPOT up to 1.03B parameters) demonstrate that Polyphase Inference consistently improves zero-shot transfer, supports parameter-efficient fine-tuning, and offers favorable time-memory trade-offs. The method is limited to regular grids with integer resolution gaps, with future work targeting irregular geometries, point clouds, and other modalities such as vision, audio, and time series.
Critical strengths
- Polyphase Inference is simple, requires no retraining, and works across architectures (FNO, UNet, Transolver, LaMO) and scales to large pretrained models (DPOT up to 1.03B parameters). It reopens a problem the field had largely abandoned and shifts the research question from "how do we train resolution-robust models?" to "how do we present high-resolution data to a fixed low-resolution model?"
- The paper does not offer decorative theory. Proposition 1 explains why direct inference fails (layer-wise aliasing propagation), Proposition 2 explains why compression cannot work in principle (irreducible error), and Propositions 3–4 explain when the proposed method works and what remains to be fixed. The error decomposition directly motivates the design of Polyphase Fine-Tuning, with in-component error justifying PEFT and cross-component error justifying the lightweight depthwise convolutions. The honesty about limitations (regular grids, integer resolution gaps, approximate marginal invariance) strengthens rather than weakens credibility.
I personally think this is a strong paper that makes a genuine conceptual contribution to an important problem. The method is simple, theoretically grounded, empirically validated across a wide range of settings, and practically deployable.
Critical weaknesses
- The paper's theoretical foundation rests on component-wise marginal invariance (Definition 1, Proposition 3). The assumption that each decomposed polyphase component's input-output distribution matches the pretraining distribution. The paper itself acknowledges this does not always hold, noting that "when exact translation invariance does not hold, spatial Sobolev regularity controls differences between phase distributions; comparison with additionally contains the mismatch of a reference phase with the pretraining distribution." The paper's own Appendix F.5 formalizes this as an approximate condition governed by an unmeasurable term , described as "separately measures the reference-phase mismatch with pretraining." This limits the method's practical utility and leaves the theoretical guarantee as a post hoc explanation rather than an actionable criterion.
- Proposition 4 decomposes the excess risk into in-component and cross-component errors. The paper correctly identifies that cross-component error requires "exploiting interactions across components" and proposes lightweight depthwise convolutions before decomposition and after recomposition, motivated by Equation 15 showing that fine-grid convolution induces structured polyphase interactions. However, the paper provides no theoretical analysis of whether these convolutions actually reduce the cross-component error, nor any ablation isolating their contribution quantitatively.
Questions for authors
- The paper positions Polyphase Inference as an alternative to the multi-resolution training advocated by Sakarvadia et al. (2026), but never directly compares the two under matched data budgets. Could the authors clarify whether, in the 1% and 10% few-shot settings, a multi-resolution training baseline was considered? This matters because the paper's practical recommendation ("use Polyphase instead of retraining") is only justified if Polyphase matches or exceeds the alternative at comparable cost. If multi-resolution training with the same 1% high-resolution data outperforms Polyphase + PEFT, the paper's central practical claim would need qualification.
- The polyphase decomposition produces components. For 3D problems with , this is 64 forward passes. Could the authors clarify whether the method was tested in any 3D setting, and whether the wall-clock or memory advantage over direct high-resolution inference persists as and grow? The efficiency claims in Section 5.5 are based on 2D data; extrapolation to 3D is not obvious.
- The paper positions Polyphase Inference as an alternative to the multi-resolution training advocated by Sakarvadia et al. (2026), but never directly compares the two under matched data budgets.
I recommend acceptance, with the expectation that the authors will clarify the questions raised above in the camera-ready version.
5 comments
Part 2/2 - Response to Reviewer Zro8
Q1, Q3: Comparison with multi-resolution training advocated by Sakarvadia et al. (2026).
We did not include this baseline in the original submission because our setting assumes an existing coarse-resolution pretrained model. However, we agree that multi-resolution training is an important alternative when additional high-resolution training data is available.
We have conducted additional experiments comparing multi-resolution FNO training with PPFT under the same high-resolution data budgets (1% and 10%) and under a favorable budget for multi-resolution training (25% of data for each resolution). The results are reported in the table below, where we compare multi-resolution training against the best PPFT configuration, with or without adapters.
With 1% high-resolution data, PPFT outperforms multi-resolution training in all 6 configurations across Darcy, Burgers, and Reaction–diffusion. With 10% data, PPFT achieves lower errors in 3 out of 6 configurations, while multi-resolution training remains competitive, particularly on Reaction–diffusion.
Overall, PPFT achieves better results in 9 out of 12 configurations. We report a table for comparison here, and we added more detailed results in Tab.32-33.
| Dataset / split | Best epoch | Multi-resolution test relative L2 by resolution | Best PPFT at same % of high res. data |
|---|---|---|---|
| Darcy | 128² / 64² / 32² / 16² | 128² | |
| 25/25/25/25 | 146 | 0.0481 / 0.0423 / 0.0430 / 0.0562 | — |
| 90/0/0/10 | 146 | 0.0563 / 0.0553 / 0.0540 / 0.0520 | 0.0534 |
| 99/0/0/1 | 145 | 0.0927 / 0.0757 / 0.0727 / 0.0509 | 0.0631 |
| 0/0/99/1 | 146 | 0.0753 / 0.0399 / 0.0428 / 0.0792 | 0.0364 |
| 0/0/90/10 | 141 | 0.0539 / 0.0410 / 0.0428 / 0.0791 | 0.0355 |
| Burgers | 1024 / 512 / 256 / 128 | 1024 | |
| 25/25/25/25 | 145 | 0.0496 / 0.0469 / 0.0482 / 0.0554 | — |
| 90/0/0/10 | 146 | 0.0553 / 0.0509 / 0.0519 / 0.0550 | 0.0588 |
| 99/0/0/1 | 148 | 0.0725 / 0.0531 / 0.0537 / 0.0562 | 0.0567 |
| 0/0/99/1 | 145 | 0.0660 / 0.0451 / 0.0469 / 0.0612 | 0.0460 |
| 0/0/90/10 | 146 | 0.0519 / 0.0459 / 0.0475 / 0.0602 | 0.0457 |
| Reaction–diffusion | 128² / 64² / 32² / 16² | 128² | |
| 25/25/25/25 | 142 | 0.2428 / 0.2397 / 0.2415 / 0.3099 | — |
| 90/0/0/10 | 131 | 0.3232 / 0.3118 / 0.3090 / 0.3234 | 0.3320 |
| 99/0/0/1 | 149 | 0.4338 / 0.4093 / 0.4083 / 0.3315 | 0.3320 |
| 0/0/99/1 | 146 | 0.2247 / 0.1977 / 0.2056 / 0.3756 | 0.2210 |
| 0/0/90/10 | 135 | 0.2097 / 0.1976 / 0.2059 / 0.3757 | 0.2210 |
Q2. Efficiency and evaluation on 3D data.
We added two 3D datasets (Swift-Hohenberg and Kuramoto-Sivashinsky) generated with the solvers provided in APEBench (Koehler et al., 2024) at resolutions and . We added the results in Tab.7-8 and we included the results in Fig.3 and Fig.8 with latency and memory peak measurements. At these resolutions, polyphase inference is, roughly, as fast as the direct evaluation of FNO.
The efficiency benefits are significant at larger scale. To demonstrate this, we added experiments scaling the resolution in 1D, 2D, 3D, and 4D, using randomly generated inputs, since the analysis concerns computational efficiency rather than predictive accuracy, until reaching out-of-memory (OOM) on an NVIDIA A100 GPU with 80 GB of memory. At the largest resolutions, parallel PPI achieves speedups over direct FNO of 3.59× in 1D (50,331,648 points), 3.36× in 2D (), 2.99× in 3D (), and 3.04× in 4D (), with comparable peak memory consumption. We added these results as Tab.34-37 in App.L.1 and a visualization of the scaling trend in Fig.7.
Re: Response to Reviewer Zro8
Thank you very much for the authors' careful and conscientious response, as well as the extensive additional experiments. I truly appreciate the effort and seriousness the authors have brought to this revision.
I personally find the core idea appealing. That said, the concerns raised by the other reviewers have also caught my attention, and I would very much like to understand how the authors view and plan to address those points. I would also kindly encourage the authors to directly and fully respond to Reviewer b6zW's second-round comments.
At this stage, and with genuine regret, I am afraid I must maintain my current assessment unchanged.
Thank you again for the authors' hard work and thoughtful response.
Part 1/2 - Response to Reviewer Zro8
First of all, thank you very much for the time and care you have invested in reviewing our paper. We sincerely appreciate it.
We agree that the condition involving W2(ν0,μr) has limited utility as a practical criterion: it cannot generally be verified from low-resolution training data alone, and estimating it reliably from high-dimensional samples can be difficult even when high-resolution data are available.
The purpose of this condition is to clarify the assumptions needed to transfer low-resolution accuracy to high-resolution inference. It shows how distributional mismatch affects the theoretical guarantee, but does not provide a practical way to check whether the method will work. Even though both resolutions describe the same physical system, their distributions are not necessarily close.
Our experiments show that Polyphase Inference works well across several PDE benchmarks, but they do not prove that the Wasserstein condition holds. We therefore see the theory as an explanation of when the method can be expected to work, rather than a guarantee that it will work in every setting
We have added a brief clarification on this point in Appendix F.5 of the revised manuscript.
Sec. 4.5 provides a theoretical motivation for the proposed convolutions: it shows that convolution on the high-resolution grid induces interactions between polyphase components, allowing each component to incorporate information from the others.
For fixed polyphase decomposition and recomposition operators, Proposition 4 shows that the cross-component error term is independent of the pretrained model. Hence, adapting the pretrained model alone cannot reduce this term. The proposed convolutions introduce interactions between polyphase components and make the decomposition and recomposition operators trainable, thereby providing a means to modify this error term. However, this does not guarantee that optimization will reduce it.
To isolate the contribution of the learned convolutions to learning decomposition and recompositions, we conducted an additional ablation study comparing PPFT + Adapter with Frozen PPI + Adapter, in which the polyphase decomposition and recomposition remain fixed and only the adapters are trained. Both methods use the same pretrained checkpoints and fine-tuning data.
The complete results are reported in Tab. 38-41 of App. L.2, the following table reports the relative L2 errors on Darcy flow:
First, Frozen PPI + Adapter consistently improves over both zero-shot PPI and adapters alone, showing that fine-tuning the pretrained model while retaining the fixed polyphase decomposition is already beneficial. Compared with zero-shot PPI, Frozen PPI + Adapter reduces the error by 5.1–10.3% with 1% of fine-tuning data and 5.1–13.3% with 10%.
Learning the decomposition and recomposition operators provides further improvements. On Darcy flow, PPFT + Adapter consistently outperforms Frozen PPI + Adapter across all pretrained resolutions and both fine-tuning budgets. The relative error is reduced by 24.2%, 4.9%, and 10.1% with 1% of fine-tuning data, and by 37.0%, 8.3%, and 12.0% with 10%, for pretrained resolutions of 16×16, 32×32, and 64×64, respectively.
On the other three datasets, the difference between Frozen PPI + Adapter and PPFT + Adapter is small. This suggests that fixed polyphase decomposition already captures most of the available benefit in these settings, leaving limited room for further improvements from learning the decomposition and recomposition operators.