- Rating
- 3(+1)
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 3
Read review
Paper summary
does a model trained only on eulerian fields support closed loop particle rollout? TLO combines a local global encoder, a latent processor, and a coordinate conditioned decoder to answer this questions. The same model forecasts fields on a fixed grid and supplies velocities at moving particle positions. The evaluation covers five field prediction and three particle rollout benchmarks, with additional experiments on decoder adaptation and prediction beyond the supervised grid support.
Critical strengths
-
The Eulerian results are strong. TLO achieves the lowest reported error on all five benchmarks in Table 2, and the ablations support combining local and global information. This is useful evidence that the proposed representation works well for field forecasting.
-
I found the distinction between Eul, Ref, and Path informative. The results show that better grid predictions do not necessarily produce better particle trajectories, which is an important point for evaluating these models. I also appreciate that the authors keep Eulerian-only training and decoder adaptation seperate.
Critical weaknesses
-
My main hesitation is the contribution beyond existing coordinate-query operators. UPT and LNO already separate latent processing from decoding at requested spatial locations. TLO adds a local-global patch representation and a Transolver-inspired processor, but Table 3 only evaluates these choices through Eulerian error. It remains unclear which components actually help with moving-particle inference. The explicit transfer evaluation is useful, and I do not assume that prior work tested exactly this protocol. Still, I would find the contribution more convincing if the architectural choices were connected more directly to improvements in Ref and Path.
-
The particle results leave this question open. In Table 4, Transolver has lower Ref and Path errors on Burgers3D and SEA, despite TLO’s better Eulerian scores. Direct decoding and high-resolution bilinear interpolation also perform very similarly in Table 12. The beyond-grid comparision could provide a stronger distinction but interpolation is simply undefined after grid exit, and other coordinate decoders are not evaluated there. This shows that TLO can continue producing predictions, but does not establish a particular accuracy advantage. The authors acknowledge these limitations, which I appreciate. Given the overlap with existing methods, however, I am not yet convinced that the demonstrated benefit for particle rollout is substantial enough for acceptance.
Questions for authors
-
Additional evidence: which architectural components are responsible for the particle-transfer performance? Could the local/global and processor ablations also report Ref and Path? Particle results for GINO, which is already included in the Eulerian comparison, would also help assess the contribution.
-
Clarification: does Table 12 use an Eulerian-only or decoder-adapted checkpoint? Are direct-decoding versus interpolation results available from the same Eulerian-only checkpoint across all three particle benchmarks, using identical integration and boundary handling?
-
Additional evidence: how does TLO compare beyond the grid support with another coordinate-query operator, such as UPT or LNO, or with interpolation equipped with an explicit extrapolation rule? A comparison under the same Eulerian-only supervision, reporting outside-region errors separately, could materially change my assessment.
2 comments
- EO readout comparisons
The original Table 12 used a decoder-adapted checkpoint. We now provide EO Direct comparisons on DAM2D, Burgers3D, and SEA. For each dataset, Direct and Interp use the same EO checkpoint and Eulerian feedback, with identical integration and boundary handling as specified in Table 1. Only the particle-velocity readout changes. In the resolution experiments, we vary only the sampling density of the interpolation grid. These results show how readout resolution affects particle prediction under the same model and particle rollout settings.
| Dataset | Resolution | Interp Ref | Direct Ref | Interp Path | Direct Path |
|---|---|---|---|---|---|
| DAM2D | 32×64 | 2.34E-03 | 2.41E-03 | 3.98E-03 | 3.72E-03 |
| DAM2D | 16×32 | 3.29E-03 | 2.41E-03 | 6.05E-03 | 3.72E-03 |
| DAM2D | 8×16 | 1.13E-02 | 2.41E-03 | 2.49E-02 | 3.72E-03 |
| DAM2D | 4×8 | 3.56E-02 | 2.41E-03 | 6.13E-02 | 3.72E-03 |
| Burgers3D | 32×32×32 | 6.19E-05 | 6.17E-04 | 2.02E-03 | 3.01E-03 |
| Burgers3D | 16×16×16 | 1.93E-04 | 6.17E-04 | 2.45E-03 | 3.01E-03 |
| Burgers3D | 8×8×8 | 1.16E-03 | 6.17E-04 | 3.06E-03 | 3.01E-03 |
| Burgers3D | 4×4×4 | 3.21E-03 | 6.17E-04 | 4.96E-03 | 3.01E-03 |
| SEA | 180×300 | 2.08E-02 | 3.00E-02 | 8.70E-03 | 8.99E-03 |
| SEA | 90×150 | 2.13E-02 | 3.00E-02 | 8.75E-03 | 8.99E-03 |
| SEA | 60×100 | 2.22E-02 | 3.00E-02 | 8.81E-03 | 8.99E-03 |
| SEA | 45×75 | 2.34E-02 | 3.00E-02 | 8.87E-03 | 8.99E-03 |
| SEA | 30×50 | 2.67E-02 | 3.00E-02 | 9.00E-03 | 8.99E-03 |
| SEA | 18×30 | 3.57E-02 | 3.00E-02 | 9.32E-03 | 8.99E-03 |
- Beyond-grid comparisons
We have also added EO beyond-grid comparisons with coordinate-query operators and grid-based baselines equipped with explicit extrapolation.
We have strengthened Section 6.3 with both coordinate-query and grid-based baselines. We retrained LNO, GINO, and Transolver using the same in-domain data and training settings as TLO. LNO, GINO and TLO use direct coordinate queries to predict velocities outside the grid.
Transolver uses bilinear interpolation inside the grid. Outside, constant extrapolation extends the boundary values, while linear extrapolation uses the slope between the two outermost grid layers. These baselines allow a direct comparison of beyond-grid accuracy.
- TLO achieves the lowest outside-region Path error under EO supervision. Compared with LNO and GINO, TLO reduces Outside@20 Path error by 85.7% and 83.2%, respectively. Transolver’s errors vary substantially with the extrapolation rule: constant extrapolation gives it the lowest outside-region Ref, but its Path remains higher than TLO’s.
- The extrapolation rule substantially affects Transolver’s errors, and the distinction between Ref and Path. TLO achieves more accurate closed-loop particle prediction, demonstrating its advantage in Eulerian-to-Lagrangian transfer.
- Decoder adaptation (DA) provides the largest improvements for TLO, reducing outside-region Ref and Path errors by 71.7% and 42.0%, respectively. TLO-DA achieves the lowest errors among the adapted models and outperforms both Transolver extrapolation baselines, further demonstrating the adaptability of TLO's learned representation for beyond-grid particle prediction.
Together with the component ablations, these results demonstrate that TLO’s local–global design supports effective Eulerian-to-Lagrangian transfer, achieving accurate closed-loop particle prediction even beyond the Eulerian output grid.
We hope these additional results clarify our contributions and address the reviewer's concerns.
| Training Setting | Model | Read-out | In-grid@10 Ref ↓ | In-grid@10 Path ↓ | Outside@20 Ref ↓ | Outside@20 Path ↓ | Full@20 Path ↓ |
|---|---|---|---|---|---|---|---|
| EO | LNO | Direct | 1.02E-02 | 3.78E-02 | 1.56E-01 | 4.23E-01 | 8.05E-02 |
| EO | GINO | Direct | 1.32E-02 | 1.51E-02 | 1.64E-01 | 3.60E-01 | 9.59E-02 |
| EO | Transolver | Linear | 8.82E-03 | 1.57E-02 | 2.89E-01 | 2.50E-01 | 4.94E-02 |
| EO | Transolver | Constant | 8.82E-03 | 1.57E-02 | 9.12E-02 | 2.37E-01 | 4.88E-02 |
| EO | TLO | Direct | 3.22E-04 | 1.89E-04 | 1.35E-01 | 6.05E-02 | 4.48E-03 |
| DA | LNO | Direct | 1.02E-02 | 3.78E-02 | 1.48E-01 | 4.23E-01 | 8.04E-02 |
| DA | GINO | Direct | 1.32E-02 | 1.51E-02 | 1.14E-01 | 3.59E-01 | 9.59E-02 |
| DA | TLO | Direct | 3.22E-04 | 1.89E-04 | 3.82E-02 | 3.51E-02 | 2.51E-03 |
Thank you for the careful review and helpful suggestions. We appreciate your recognition of TLO’s Eulerian forecasting performance and the value of distinguishing Eul, Ref, and Path.
We agree that TLO shares query-independent latent processing and coordinate-conditioned decoding with UPT/LNO. Our contribution, however, is to formulate zero-shot Eulerian-to-Lagrangian transfer, unifying field prediction and particle simulation, and introduce Eul/Ref/Path to measure cross-representation transfer. This combines a controlled, systematic study of EO transfer with an architecture designed for particle rollout.
To address transfer task, TLO couples a Local Perceiver with global aggregation through domain decomposition and uses attention and FFN to capture latent flow dynamic. Our architectural contribution lies in this particular combination of local–global representation and latent processing.
As you pointed out, the Eul-only ablations in Table 3 do not establish which components help particle prediction. We have therefore added Ref/Path results and extended the ablations to DAM, NS, and TGV. The results show that the Local Perceiver is essential for accurate particle rollout in TLO: compared with the variant without Local component, the full model reduces Path error by 35.1%, 67.7%, and 93.5%, respectively. Removing Global or FFN also increases Ref and Path. These results directly connect the architectural choices to particle-transfer performance. In the tables, Improvement denotes the error reduction of the full model relative to each ablated variant: : $\text{Improvement}(%)= (E_{\text{Baseline}}-E_{\text{Full}} )/E_{\text{Baseline}}\times100%$.
We have added GINO’s particle-prediction results and completed the NS2D and native TGV3D evaluations, extending particle evaluation to all five datasets. The expanded comparison shows TLO achieves strong Ref and Path results on these flow datasets.