- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 1
- Soundness
- 2
Read review
Paper summary
The proposed methods tackles two problems that occur in particle fluide simulations with neural methods:
- error accumulations of rollouts and
- high computational cost of message passing if graph neural networks are used. Their main trick is to keep a grid of hidden/latent features between timesteps. The model then moves information from particles to the grid, updates the grid, then moves information back to the particles. I utilizes a graph network before and after the grid step. A second trick (PGS), skips the graph networks on some steps, via the help of some smaller networks.
Critical strengths
The presented results are strong. Table 1 shows that CoLEP is beating the baselines on all of the rollout metrics across the four datasets.
Critical weaknesses
I am missing a discussion/comparison section for other latent-space methods. There have been other methods that are fairly similar, to the proposed method. UPT (Alkin et al., 2024) encodes physical data, runs dynamics in latent space, and decodes predictions. GINO (Li et al., 2023) moves irregular-point data onto a regular latent grid, processes it, and maps it back. MultiScale MeshGraphNets also uses coarse representations to make learned simulation cheaper. CoLEP is not the same: its fixed grid remembers information while particles keep moving, and the two update each other at every step. But the paper should explain this difference much more clearly. UPT (or the followup AB-UPT (Alkin et al., 2025)) is worth considering as a baseline if it can be adapted to the same particle rollouts.
Questions for authors
- Table 1, how many training seeds were used for each method and dataset? Can you show variation or uncertainty for the full-rollout scores, not just the error curves?
- Can you give a side-by-side table of parameter counts, training steps, training time, and inference setup for CoLEP and each baseline? Appendix B.5 gives some training details, but it is hard to judge whether the budgets are matched.
- Why is the repeated particle-grid update better than rolling forward in a latent space as UPT does? Can you compare against UPT on a matching particle task?
- Can you compare PGS against a smaller or shallower full-step model at about the same inference time? This would test whether skipping graph steps is better than simply using a cheaper model.
6 comments
Responses to Questions
Response to Q1: Multi-Seed Benchmark Results
Thanks for pointing this out. In the initial submission, the results reported in Table 1 were obtained using a single training seed. To properly quantify variation and uncertainty, we have retrained and evaluated all methods across 3 independent random seeds for each dataset in the revised manuscript.
Specifically, we have added the complete 3-seed results in Appendix D of the revised manuscript, reporting the mean and standard deviation for every metric, and we also provide the full table below for convenience.
| Dataset | Model | MSE₂₀ ↓ | KE–MSE₂₀ ↓ | MSE_full ↓ | KE–MSE_full ↓ | Sinkhorn_full ↓ |
|---|---|---|---|---|---|---|
| DAM-2D | GNS | (4.47 ± 0.66) × 10⁻⁵ | (3.29 ± 0.97) × 10⁻⁵ | (3.17 ± 2.57) × 10⁻¹ | (1.39 ± 1.29) × 10⁻² | (4.34 ± 3.19) × 10⁻² |
| NeuralMPM | (5.27 ± 0.70) × 10⁻⁵ | (3.53 ± 0.31) × 10⁻⁴ | (2.82 ± 0.40) × 10⁻² | (5.66 ± 1.84) × 10⁻⁴ | (9.48 ± 6.15) × 10⁻⁴ | |
| CoRGI | (1.76 ± 0.23) × 10⁻⁵ | (2.14 ± 1.96) × 10⁻⁵ | (8.26 ± 5.27) × 10⁻² | (2.21 ± 1.99) × 10⁻³ | (4.83 ± 4.93) × 10⁻³ | |
| CoLEP (Ours) | (1.31 ± 0.01) × 10⁻⁵ | (2.66 ± 0.15) × 10⁻⁶ | (2.16 ± 0.10) × 10⁻² | (3.35 ± 0.78) × 10⁻⁵ | (2.52 ± 0.26) × 10⁻⁴ | |
| TGV-2D | GNS | (7.04 ± 0.28) × 10⁻⁶ | (2.40 ± 0.26) × 10⁻⁷ | (5.74 ± 0.04) × 10⁻⁴ | (2.23 ± 0.23) × 10⁻⁷ | (9.85 ± 2.56) × 10⁻⁷ |
| NeuralMPM | (2.37 ± 0.04) × 10⁻⁵ | (6.01 ± 2.21) × 10⁻⁶ | (9.59 ± 0.41) × 10⁻⁴ | (4.13 ± 1.55) × 10⁻⁶ | (8.99 ± 1.18) × 10⁻⁶ | |
| CoRGI | (5.91 ± 0.33) × 10⁻⁶ | (1.44 ± 0.30) × 10⁻⁷ | (4.81 ± 0.21) × 10⁻⁴ | (1.39 ± 0.58) × 10⁻⁷ | (1.14 ± 1.05) × 10⁻⁶ | |
| CoLEP (Ours) | (3.55 ± 0.07) × 10⁻⁶ | (5.96 ± 0.85) × 10⁻⁸ | (3.23 ± 0.06) × 10⁻⁴ | (6.09 ± 1.23) × 10⁻⁸ | (2.04 ± 0.82) × 10⁻⁷ | |
| Moving Obstacle-2D | GNS | (9.78 ± 0.62) × 10⁻⁶ | (7.24 ± 0.89) × 10⁻⁷ | (2.03 ± 1.15) × 10⁻² | (2.84 ± 0.79) × 10⁻⁵ | (2.26 ± 1.70) × 10⁻² |
| NeuralMPM | (2.84 ± 0.21) × 10⁻⁵ | (1.38 ± 0.35) × 10⁻⁶ | (1.70 ± 0.10) × 10⁻³ | (2.00 ± 0.24) × 10⁻⁶ | (9.03 ± 0.46) × 10⁻⁴ | |
| CoRGI | (6.68 ± 0.19) × 10⁻⁶ | (4.07 ± 1.02) × 10⁻⁷ | (3.39 ± 0.26) × 10⁻³ | (4.23 ± 0.35) × 10⁻⁶ | (2.39 ± 0.51) × 10⁻³ | |
| CoLEP (Ours) | (1.61 ± 0.06) × 10⁻⁶ | (1.52 ± 0.01) × 10⁻⁷ | (8.92 ± 0.96) × 10⁻⁴ | (5.58 ± 1.13) × 10⁻⁷ | (2.47 ± 0.08) × 10⁻⁴ | |
| Water-3D | GNS | (5.63 ± 0.41) × 10⁻⁵ | (4.97 ± 2.30) × 10⁻⁴ | (8.75 ± 2.29) × 10⁻² | (9.66 ± 6.52) × 10⁻³ | (1.10 ± 0.75) × 10⁻¹ |
| CoRGI | (4.00 ± 0.30) × 10⁻⁵ | (3.28 ± 0.13) × 10⁻⁴ | (6.79 ± 1.92) × 10⁻² | (3.27 ± 1.88) × 10⁻³ | (6.53 ± 5.56) × 10⁻² | |
| CoLEP (Ours) | (3.72 ± 0.15) × 10⁻⁶ | (1.91 ± 0.68) × 10⁻⁵ | (1.42 ± 0.02) × 10⁻² | (3.87 ± 0.69) × 10⁻⁵ | (6.28 ± 2.65) × 10⁻³ |
Table R1-1: Quantitative comparison of 20-step and full-rollout performance across all benchmark datasets. Results are reported as mean ± standard deviation across 3 independent random seeds, with the best scores highlighted in bold.
Response to Q2: Model Costs and Training Protocols
To clarify the computational budgets and evaluation protocols across methods, we report the exact parameter counts across datasets in Table R2-1, training budgets and wall-clock training times on DAM-2D in Table R2-2, as well as CoLEP's training schedules and grid configurations in Tables R2-3 and R2-4, followed by a summary of our unified inference setup.
Table R2-1 reports the exact parameter counts for CoLEP and each baseline across the benchmark datasets. Note that NeuralMPM is restricted to 2D domains and is therefore omitted on Water-3D.
| Dataset | CoLEP Full | GNS | NeuralMPM | CoRGI |
|---|---|---|---|---|
| DAM-2D | 3,407,338 | 1,211,794 | 1,117,200 | 17,437,586 |
| TGV-2D | 3,966,178 | 1,211,538 | 1,117,200 | 17,437,330 |
| MovingObstacle-2D | 3,373,538 | 1,212,050 | 1,116,418 | 4,369,170 |
| Water-3D | 8,481,763 | 1,213,203 | — | 11,318,451 |
Table R2-1: Exact parameter counts across benchmark datasets.
Because our training runs were distributed across NVIDIA A100 and RTX 4090 GPUs, wall-clock training times cannot be directly aligned across all datasets. Table R2-2 therefore only reports the training times on DAM-2D, where all four methods were trained on the same NVIDIA A100 GPU. Across all experiments, we followed the official implementations and standard training protocols for each baseline so that all models reached full convergence. Specifically, we used the LagrangeBench implementation of GNS and trained each seed for 500,000 optimizer updates across all datasets. NeuralMPM and CoRGI were trained using their official rollout protocols, using 101,000 updates with a batch size of 128 for NeuralMPM, and 500,000 updates with a batch size of 4 for CoRGI. CoLEP employs multi-step rollout curriculum supervision that progressively scales from 2 to 20 steps, requiring 30,000 optimizer updates on DAM-2D while achieving a wall-clock training time comparable to CoRGI.
| Method | Batch Size | Training Steps per Seed | Training Time (h) |
|---|---|---|---|
| CoLEP | 4 | 30,000 | 28.99 |
| GNS | 1 | 500,000 | 7.11 |
| NeuralMPM | 128 | 101,000 | 19.86 |
| CoRGI | 4 | 500,000 | 27.12 |
Table R2-2: Training budgets and wall-clock training time per seed on DAM-2D.
For completeness, Table R2-3 details CoLEP's total training steps and progressive rollout transition ranges across all five datasets, and Table R2-4 lists the corresponding core and computational grid resolutions along with particle interaction radii.
For inference setup, all methods are evaluated on the same single A100-SXM4-40GB GPU with a batch size of 1. GNS, CoRGI, and our CoLEP variants perform single-step autoregressive rollouts initialized from 6 historical frames, whereas NeuralMPM predicts 8 frames per forward pass normalized over the 20-step horizon.
| Dataset | Total training steps per seed | R2 → R4 transition | R4 → R8 transition | R8 → R16 transition | R16 → R20 transition | Fixed R20 training range |
|---|---|---|---|---|---|---|
| DAM-2D | 30,000 | 1,001–1,500 | 2,751–3,500 | 5,251–6,250 | 8,501–10,000 | 10,001–30,000 |
| TGV-2D | 50,000 | 2,779–4,167 | 7,640–9,722 | 14,584–17,361 | 23,612–27,778 | 27,779–50,000 |
| MovingObstacle-2D | 18,000 | 1,001–1,500 | 2,751–3,500 | 5,251–6,250 | 8,501–10,000 | 10,001–18,000 |
| Water-3D | 39,500 | 10,001–15,000 | 24,251–25,000 | 26,751–27,750 | 30,001–31,500 | 31,501–39,500 |
| TGV-3D | 20,000 | 1,001–1,500 | 2,751–3,500 | 5,251–6,250 | 8,501–10,000 | 10,001–20,000 |
Table R2-3: CoLEP training steps and progressive rollout curriculum schedules across datasets.
| Dataset | Core grid | Computational grid | Particle interaction radius |
|---|---|---|---|
| DAM-2D | 64 × 160 | 72 × 168 | 0.029 |
| TGV-2D | 64 × 64 | 64 × 64 | 0.029 |
| MovingObstacle-2D | 64 × 64 | 72 × 72 | 0.015 |
| Water-3D | 32 × 32 × 32 | 36 × 36 × 36 | 0.035 |
| TGV-3D | 32 × 32 × 32 | 32 × 32 × 32 | 0.46 |
Table R2-4: CoLEP grid resolutions and particle interaction radii across datasets.
Thanks for the careful assessment and constructive feedback. We appreciate your recognition of CoLEP's strong experimental results across all four benchmarks.
We would like to reiterate our two main contributions: (1) a persistent Eulerian latent field that co-evolves with Lagrangian particles, enabling spatial information to be accumulated and reused across time steps to improve long-horizon prediction; and (2) Periodic GNN Skipping (PGS), which leverages this persistent field to reduce expensive particle interactions while maintaining prediction accuracy.
We address your concerns and provide additional experimental evidence below.
Response to W1: Latent-Space Method Comparison
Thanks for your constructive suggestion. We fully agree that expanding the related work section makes our literature review more comprehensive and that clarifying the differences between CoLEP and UPT [1] is necessary. To address this, we revised the paper in two aspects. First, we expanded the Related Work section in the main text to introduce UPT [1], GINO [2], and MultiScale MeshGraphNets [3]. Second, we added a comparison section in Appendix F to discuss the architectural differences between CoLEP and related latent-space methods, along with experimental comparisons against a representative method of latent method UPT.
Regarding whether UPT is a suitable baseline for this task, we examine this question from both architectural design and experimental validation. First, from an architectural perspective, alongside dedicated particle-based simulators such as GNS and CoRGI, our CoLEP focuses on Lagrangian particle simulation and requires accurately predicting individual particle trajectories over long autoregressive rollouts. In contrast, UPT is primarily designed as a general-purpose framework for continuous field modeling and velocity field prediction. Because of this fundamental difference in task formulation, UPT's architecture is not tailored to capture particle interactions that are essential for preventing error accumulation in particle simulation tasks.
In Response to Q3, we provide a detailed architectural analysis explaining how these design choices affect its behavior on pure particle simulation tasks. To further validate this empirically, we evaluated UPT on several benchmark datasets, and the results confirm that while it can be adapted to particle rollouts, it is less competitive than dedicated neural particle simulators such as GNS and CoRGI. Detailed experimental settings, quantitative results, and further analysis are provided in Response to Q3 and have been updated in Appendix F.
References
[1] Benedikt Alkin, Andreas Fürst, Simon Schmid, Lukas Gruber, Markus Holzleitner, and Johannes Brandstetter. Universal Physics Transformers: A Framework For Efficiently Scaling Neural Operators. Advances in Neural Information Processing Systems, 37, 2024.
[2] Zongyi Li, Nikola Kovachki, Chris Choy, Boyi Li, Jean Kossaifi, Shourya Otta, Mohammad Amin Nabian, Maximilian Stadler, Christian Hundt, Kamyar Azizzadenesheli, and Anima Anandkumar. Geometry-Informed Neural Operator for Large-Scale 3D PDEs. Advances in Neural Information Processing Systems, 36, 2023.
[3] Meire Fortunato, Tobias Pfaff, Peter Wirnsberger, Alexander Pritzel, and Peter Battaglia. MultiScale MeshGraphNets. 2nd AI4Science Workshop at ICML, 2022.