- Rating
- 3
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
The paper makes linear attention skew-symmetric, so pairwise latent fluxes cancel and any linearly projected quantity is conserved exactly at O(N) cost. It adds a conservative grid-based MeshConv for near-field interactions and a sign-split flux limiter that enforces bounds such as non-negative concentration. On 2D TGV and lid-driven cavity, the method conserves mass and momentum almost exactly and scales to about 38× more particles than graph-based baselines, at somewhat lower accuracy than GNS.
Critical strengths
The core idea is simple and meaningful: exact conservation by construction, linear complexity, and no graph construction. The limiter preserves both skew-symmetry and O(N) evaluation. The scalability gains are clear and practically useful.
Critical weaknesses
Closely related baselines are missing: Horie & Mitsume (2024), which uses antisymmetric messages for conservation, and Transolver (Wu et al., 2024), a linear-complexity transformer for PDEs. Both are cited but not compared. The evaluation is limited to two small 2D cases with 2,500 particles. The 3D result is qualitative only, and accuracy is below GNS. Quantitative results on a standard benchmark such as LagrangeBench, or at least 3D error against ground truth, would strengthen the paper considerably.
Questions for authors
Is the limiter used during training or only at inference? Also, which Table 2 configuration matches the full model in Table 1? The concentration losses don’t line up.
Main Reply
A: Thank you for pointing out the missing definitions. First, for a scalar x, we define x+=max(x,0) and x−=max(−x,0). Both parts are nonnegative, and x=x+−x−. The same convention applies entrywise to L and R.
Second, τi=Δt/Vi, where Δt is the transport time step and Vi=mi/ρ0 is the reference volume of particle i. Dividing both sides of Eq. (13) by Vi gives this factor, which converts a flux into a change in the particle state over one step.
Third, in Appendix B.1, pk denotes the locally frozen scalar attention coefficient for the interaction from grid point i−k to grid point i, where k is a grid offset. At L194, pi=WFzFi denotes the momentum vector of particle i.
A: The reported experiments use a fixed number of Picard-style latent updates with the hard contraction condition left unenforced, since Hard contraction constraints can limit the expressiveness of the learned transport operator (Baker et al., ICML 2023) which is also verified in our early attempts. Therefore, the Picard updates is downgraded to a design intention and inductive bias in a pipeline level. Its effectiveness is verified by Fig. 5(b) where iterations with proper times help to improve the training dynamics. We have clarified this in the updated draft. Furthermore, such update rule is optional since this has nothing to do with our two main claims: conservation and scalability.
A: We summarize our response into two points as it is a bit lengthy. We will cover these implementation details in our work.
A: GNS achieves lower fluid and concentration prediction losses on both datasets. Our model runs approximately 3.2× as fast as GNS on 2D-TGV and 3.6× as fast on 2D-LDC. It also achieves lower fluid and concentration prediction losses than SFBC and FluidFormer on both datasets and has the lowest runtime in Table 1. We propose describing these results as a favorable accuracy–efficiency trade-off in the revision.
The experiments in Fig. 3b assess hardware scalability through runtime and peak memory. We evaluate rollout consistency under denser particle sampling against a reference interpolated from the coarse trajectory. For 2D TGV, we use a split-and-interpolate scheme to increase the particle count from 2,500 to 1.715M and run the model trained at the original particle count without retraining.
The table below reports position MSE ( Lfluid ) and concentration MSE (Lconc). The revised manuscript also includes the corresponding multi-resolution rollout visualizations in Appendix D (Fig. 6).
All baselines run out of GPU memory at 1.715M particles, which precludes evaluation at this scale.
A: The full TGV rollout animation appears in the README of the anonymous code repository which also includes side-by-side rollout GIFs comparing our model with the baselines on 2D-TGV and 2D-LDC, along with the full 3D dam-break rollout. Visual inspection of the 3D rollout suggests mild spatial smoothing whose severity appears broadly similar over time. Deterministic MSE training may contribute to this effect, particularly in isolated splashes and fine-scale particle structure.