- Rating
- 3
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
The paper proposes multiphysics post-training: jointly adapting separately pretrained single-physics surrogates while preserving their physical interfaces. It introduces EquiPT, which supervises coupled equilibria via implicit differentiation, and Unrolled-K, which trains through a fixed number of field exchanges. A local error analysis characterizes how module errors propagate through coupling. Experiments on neutron–thermal coupling, Gray–Scott reaction–diffusion, and fluid–structure interaction show improved accuracy over frozen assembly and module-wise training, plus analyses of deployment behavior and computational trade-offs.
Critical strengths
The paper tackles a practical problem, namely the mismatch between training modules in isolation and deploying them coupled, and offers a modular solution that preserves physical interfaces, giving a useful way to adapt existing single-physics surrogates to coupled prediction. The evidence is solid: results across three distinct physical systems, including matched-data module-wise continued-training controls, show gains beyond just more data exposure, and the deployment-depth and computational analyses make clear when equilibrium training or finite unrolling is the better choice.
Critical weaknesses
EquiPT’s methodological novelty beyond established DEQ training and related work such as DEQ-MPI could be clearer. The validation/test split also needs clarification: although training and held-out trajectories are separated, it is unclear whether validation and test sets are trajectory-disjoint, given the overlapping GS windows. Finally, the spectral diagnostics do not disentangle module-bias reduction from changes in directional error amplification.
Questions for authors
See weaknesses
4 comments
Official Response to Reviewer abY2 (2)
Weakness 2: The validation/test split also needs clarification: although training and held-out trajectories are separated, it is unclear whether validation and test sets are trajectory-disjoint, given the overlapping GS windows.
Response:
We thank the reviewer for pointing this out. After careful check, we find that there is indeed a mistake of overlapping frames between validation and test sets of the GS dataset in our original submission. To completely eliminate this, we designed a new split and re-ran all training, checkpoint selection, and evaluation for GS.
In the new split of the 25 GS trajectories, the training data remain unchanged (16 trajectories for pre-training and 4 for post-training). The remaining 5 held-out trajectories are partitioned into 2 validation trajectories and 3 test trajectories. Then all training, validation, and test sets are mutually disjoint and share no overlapping frames. For all methods, checkpoints are selected strictly based on validation error, and the test set is used once only at evaluation.
Under this new split, we retrained all GS methods in Table 1 of the revised manuscript while keeping all other configurations identical to our previous setup, again across 3 random seeds (GenCP and M2PDE: single runs). The updated results are shown in Table Rebuttal D1, with relative sample standard deviations in parentheses. These results confirm that the trajectory-disjoint split does not change any of our conclusions. Across all methods, the system errors remain within roughly 7% of the original values, and EquiPT still achieves the lowest error across all metrics. Specifically, EquiPT reduces system error by 50.2% relative to Frozen assembly (vs. 48.7% originally) and by 52.2% relative to MW-CT trained on the same coupled data (vs. 54.2% originally).
In the revised manuscript, we have replaced the GS columns in Table 1 and the corresponding results with these retrained numbers. Due to time constraints, we have retrained only the GS results in Table 1 and the numbers derived from them. Because the new split leaves these results essentially unchanged (all within about 7%), it does not affect any analysis built on them, so the remaining GS analyses keep the original split and will be updated in the next revision. We thank the reviewer again for catching this issue and helping us improve the manuscript.
Table Rebuttal D1. GS errors under the trajectory-disjoint split. Means over three seeds (GenCP and M2PDE: single runs), relative standard deviations in parentheses.
| u | v | System | |
|---|---|---|---|
| Frozen | 0.0207 (±27.27%) | 0.0461 (±22.74%) | 0.0334 (±24.14%) |
| GenCP | 0.0199 | 0.0543 | 0.0371 |
| M2PDE | 0.2601 | 0.6499 | 0.455 |
| EquiPT w/o pre-training | 0.0189 (±3.49%) | 0.0467 (±5.63%) | 0.0328 (±4.98%) |
| MW-CT | 0.0213 (±19.14%) | 0.0483 (±15.58%) | 0.0348 (±16.61%) |
| Unrolled-1 | 0.0140 (±11.22%) | 0.0251 (±3.61%) | 0.0195 (±1.94%) |
| Unrolled-15 | 0.0112 (±14.04%) | 0.0283 (±12.79%) | 0.0197 (±13.13%) |
| EquiPT | 0.0096 (±5.85%) | 0.0237 (±4.56%) | 0.0166 (±4.93%) |
Official Response to Reviewer abY2 (3)
Weakness 3: Finally, the spectral diagnostics do not disentangle module-bias reduction from changes in directional error amplification.
Response:
We agree with the reviewer's point that the spectral radius alone cannot disentangle these two effects, which is why we treat it strictly as an accompanying diagnostic in the paper.
In fact, in Theorem 1 of the original manuscript separates the two effects directly, so we explicitly measured both. For each method, we compute three quantities: the modular bias , evaluated using reference partner fields as inputs; the equilibrium error , evaluated under each method's coupled deployment; and the amplification ratio . The results are reported in Table Rebuttal D2.
On GS and FSI, actually increases after EquiPT fine-tuning (by 77% on GS and 6% on FSI), yet EquiPT lowers the overall error by suppressing coupled error amplification (); on NT, both and decrease. It should be noted that a smaller modular bias does not by itself guarantee a smaller coupled error. Take NT as an example, MW-CT continues training on the same coupled data using the module-wise objective, thereby achieving a smaller modular bias than EquiPT, yet suffering from much stronger error amplification ( vs. ), resulting in a larger overall error. This behavior aligns with what each objective supervises: the module-wise objective never evaluates coupled predictions, whereas post-training directly supervises the coupled prediction itself. This is precisely the mismatch between module-wise objectives and coupled prediction that motivates our post-training paradigm.
At the same time, spectral diagnostics still provide valuable complementary insight. First, the spectral radius characterizes the coupling sensitivity of the learned system, and Corollary 2 of the original manuscript links it directly to error amplification by showing that the modular bias along each eigenmode is amplified by at most . Post-training substantially lowers this ceiling: on NT, the per-sample median drops from 10.13 for Frozen assembly to 3.03 for EquiPT, consistent with the empirical suppression of amplification observed above. The two diagnostics thus serve distinct roles. The spectral radius depends solely on the learned coupled system and, via Corollary 2 of the original manuscript, provides a reference-free upper bound on the gain of any individual eigenmode, whereas measures the actual amplification that the coupled system induces on the modular bias, which also depends on which modes are excited. In short, the former bounds how much the system can amplify a single mode in the worst case, while the latter measures how much it actually amplifies the realized bias. Consistent with our discussion in Section 3.2 of the original manuscript and Section 5.4 of the revised manuscript, the spectral radius is best interpreted as a supporting diagnostic alongside empirical prediction errors.
We have added this analysis to Section 5.4 and included the corresponding table in Appendix D.12 of the revised manuscript.
Table Rebuttal D2. Modular bias , equilibrium error , and amplification . uses reference partner fields as inputs; uses each method's deployment solver.
| NT | GS | FSI | |||||||
|---|---|---|---|---|---|---|---|---|---|
| B | H | H/B | B | H | H/B | B | H | H/B | |
| Frozen | 27.7 | 29.5 | 1.06 | 0.5 | 3.7 | 7.37 | 1.65 | 4.36 | 2.64 |
| MW-CT | 0.85 | 1.37 | 1.61 | 0.50 | 3.45 | 6.84 | 1.64 | 4.44 | 2.71 |
| EquiPT | 2.49 | 0.76 | 0.3 | 0.89 | 1.61 | 1.81 | 1.75 | 1.75 | 1.00 |
References
[1] Bai, Shaojie, et al. "Deep equilibrium models." NeurIPS (2019)
[2] Güngör, Alper, et al. "DEQ-MPI: A deep equilibrium reconstruction with learned consistency for magnetic particle imaging." IEEE Transactions on Medical Imaging (2024)
Thank you for the clarifications and additional analyses, which are helpful in assessing the paper. I maintain my original rating of Weak Accept.
Official Response to Reviewer abY2 (1)
We thank the reviewer for the positive assessment. We appreciate the reviewer's recognition of our modular solution to the training–deployment mismatch, which preserves physical interfaces and adapts existing single-physics surrogates, backed by solid evidence across three physical systems. The reviewer's valuable concerns on methodological novelty, the data split, and the spectral diagnostics helped us strengthen the paper, and we address each below, with revisions highlighted in blue in the revised manuscript.
Response:
We thank the reviewer for the comment. EquiPT indeed adopts the idea of differentiating through fixed points from Deep Equilibrium Models (DEQs) [1] along with existing root solvers. We do not claim this as our contribution either, as detailed below.
Standard DEQs are monolithic networks trained from random initialization, where the equilibrium is defined over internal latent features, which means existing single-physics models cannot be reused, the equilibrium has no physical ground truth for supervision, and errors cannot be decomposed across physical fields. We evaluated its most favorable variant on our problem (removing only module-wise pre-training while keeping all other components identical to EquiPT) and found that it performs markedly worse than EquiPT (Table 1 and Appendix D.11 of the revised manuscript). Meanwhile, DEQ-MPI [2] addresses the simplest setting for equilibrium methods, where a single unknown field with a known fixed system matrix is solved throughout the iterations. In contrast, our target multiphysics coupling involves multiple interacting unknown fields, where each field operator is learned with inherent approximation bias, and these modular biases mutually amplify across coupled interfaces.
As shown above, existing DEQ related methods are not designed for multiphysics coupling. Applying them directly yields poor performance and lacks theoretical characterization of cross-field error propagation. EquiPT's methodological novelty lies in embedding equilibrium differentiation as a tool within our proposed multiphysics post-training paradigm, while redesigning three core elements of equilibrium training specifically for coupled multiphysics including the equilibrium state definition, the training initialization, and the supervision objective:
We have added the discussion and relevant references to Section 2 of the revised manuscript.