- Rating
- 3(+1)
- Confidence
- 2(−1)
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper proposes TEMPO-PDE, a post-training framework for pretrained PDE predictors. It combines field-value, spectral, and spatial-difference errors and organizes trajectory supervision into primary and auxiliary objectives. For direct prediction, paired future lead times define the primary objective and the window endpoint provides auxiliary supervision. For recurrent prediction, rollout losses form the primary objective, supplemented by direct forecasts from the same observed state. The update clips the primary gradient, removes conflicting auxiliary components, and caps the auxiliary gradient norm relative to the primary gradient.
Experiments primarily use Poseidon-B and include matched training controls, component ablations, and extended-horizon forecasts. Additional studies examine score transfer to Flower-Tiny and an adapted TEMPO-AR formulation on pretrained DPOT-L. The reported results show improvements in the main Poseidon-B comparisons, while evidence for the additional benefit of the full combination is less consistent on other backbones.
Critical strengths
-
The article presents a clear post-training method. This method utilizes the existing prediction interface and introduces additional scoring and gradient processing during the training process without the need for an inference module. The main goals and auxiliary goals, as well as their processing sequence, are sufficiently clear, making the core suggestions easy to implement. This is of practical value for adjusting existing partial differential equation predictors when the target system data is limited.
-
The Poseidon-B comparison, component ablation, and long-term sequence evaluation indicate that the impact of training selection on the overall prediction quality may differ from the accuracy of the early steps. Additionally, these additional experiments also reveal limitations, such as in some cases, the configuration based on simple scoring performs better than the complete transfer version. Although these findings do not establish a complete solution that is always superior to other methods, they still provide valuable empirical information.
Critical weaknesses
-
The numerical report and evaluation scope of the article are not clear. Appendix A reports unexplainable TEMPO results (24.04/42.44, while other configurations are 0.976/1.348), and increased Flower-Tiny prediction errors without verification against the five-task list. Moreover, this appendix summarizes FinalTest144 as eight victories in twelve task-indicator cells, but does not provide the complete results; while the main comparison uses Fresh128 tested during the development process. These omissions make it difficult to judge the evaluation scope and the independent support for the reported improvement effect.
-
The methodological differences in the article are limited, and there is a lack of sufficient evidence to support a complete combination. The proposed update is closely related to MetaBalance and PEGrad, while Poseidon has achieved the reuse of time-point snapshots. The conditional projection and clipping selection of TEMPO are different, but their added value has not been fully verified. In Table 17, Score-H4 shows lower average errors than TEMPO-AR on all five DPOT-L tasks, although TEMPO-AR improved the final error on two tasks. The detailed Flower-Tiny study only tested the effect of score transfer. Overall, the evidence supports the scoring function more clearly than the innovation and added benefits of the complete combination.
Questions for authors
-
Which tasks, configurations and evaluation schemes do the TEMPO and Flower-Tiny values mentioned in the opening paragraph of Appendix A correspond to? Please compare these data with Table 4-5 and the Flower-Tiny results in Appendix E.3. If there are tasks not included in the displayed subsets, please explain the relationship between this subset and the original evaluation scope.
-
Can the author provide the complete FinalTest144 results for tasks and indicators, including the relevant comparison groups, and explain the relationship between the twelve task-indicator units and the five-task main evaluation?
-
Compared to the similar primary-secondary alternatives such as PEGrad or target-referenced gradient rescaling, what specific advantages do the proposed gradient rules offer? This request aims to confirm the current contribution rather than extending the work to a broader future benchmark.
13 comments
Part 2 of 5. Question 2: the complete FinalTest144 results
Within the comparison pool that was frozen before the test, TEMPO has the lowest error in all ten cells of the five-task main evaluation. The twelve cells of Appendix A are these ten cells plus two cells for the sixth task, ACE. The count of 8 of 12 arises when comparators that were appended later are included.
In detail, the twelve task and metric cells are the six registered tasks times the two metrics, Mean and Final. The six tasks are the five main tasks and ACE, so the five-task main evaluation corresponds to the first ten cells. All cells are computed on FinalTest144, which contains 144 test trajectories per task. Each model was evaluated on it once, and every row reported here uses two training seeds.
The comparators come from two stages. The first stage is the original pool, which was frozen before any FinalTest144 value was accessed. It contains the 13 external architectures, the residual-head control [r52] and the Poseidon-B controls registered at that time, which are [r20], [r21], [r36], [r38] and [r51]. The table gives the TEMPO entry and the strongest comparator of this pool on the five main tasks, each as Mean and Final error. Here [r38] is the per-tensor TEMPO variant, [r51] is the recurrent fine-tuning reference, and [r36] is the matched direct fine-tuning for CE-RPUI.
| Task | TEMPO entry | Strongest comparator of the original pool |
|---|---|---|
| NS-PwC | 9.121 / 12.385 | [r38] 9.255 / 12.723 |
| NS-SL | 17.373 / 28.083 | [r51] 17.947 / 28.761 |
| NS-SVS | 20.181 / 34.668 | [r51] 20.850 / 35.921 |
| FNS-KF | 9.573 / 13.638 | [r38] 9.890 / 14.242 |
| CE-RPUI | 136.024 / 206.903 | [r36] 137.598 / 208.858 |
The TEMPO entry has the lowest error in both metrics on all five main tasks, which means in all ten cells of the five-task main evaluation. This pool includes the 13 external architectures, and the best of them has 1.6 to 5.5 times the Mean error of TEMPO on the five main tasks.
The second stage appended the baseline portfolio [r13] and the DPOT-L rows. The portfolio [r13] has 9.378 / 12.689 on NS-PwC, 18.546 / 30.106 on NS-SL, 22.456 / 38.263 on NS-SVS, 9.806 / 13.995 on FNS-KF and 134.090 / 197.861 on CE-RPUI. With these comparators included, TEMPO leads in 8 of 12 cells, which is the count quoted in Appendix A. These cells are both metrics on NS-PwC, NS-SL, NS-SVS and FNS-KF, each with a Holm-adjusted p of at most 0.0093 over the twelve hypotheses.
For CE-RPUI, the submitted recipe leads on a separate confirmation cohort, and the entry on FinalTest144 is older than this recipe. The entry was produced before the four-variable rerun noted in App. B.3, and the submitted Table 1 uses the recipe of this rerun. We compared the submitted recipe with the same comparators on the confirmation cohort. This cohort has 128 trajectories per task and was set aside from unused data before the recipe was fixed. It is separate from the sets named in the general response. The comparators include the appended portfolio [r13], which on the direct tasks is the three-query supervision of TEMPO with independent sampling and ordinary summation. On this cohort the submitted recipe has a Mean and Final error of 114.585 and 166.523 against 117.018 and 169.581 for the portfolio, with a Holm-adjusted p of 0.0002 for both metrics, and the TEMPO entries lead in all ten main-task cells. On HeldOut128 the submitted recipe reaches 114.441 and 167.101, which is 5.2% and 8.0% below the fine-tuning reference at 120.759 and 181.553.
For ACE, the entry is the direct fine-tuning retained by the selection rule, as part 1 explains, and it has 23.953 / 42.498. This entry therefore compares Poseidon-B fine-tuning with other models and does not involve TEMPO post-training of Poseidon-B. Its strongest comparator in the original pool is [r52], a frozen Flower-Tiny expert with a residual head, which has 8.132 / 13.216. The strongest appended comparator is TEMPO-AR on DPOT-L with 13.146 / 20.879. The direct TEMPO recipe lowers the error of the ACE entry by 26.2% and 39.4% on Fresh128 and by 26.7% and 39.9% on HeldOut128, as part 1 shows. Within DPOT-L, TEMPO-AR is the best of the three configurations on ACE on FinalTest144.
On DPOT-L, TEMPO-AR is below the native fine-tuning in all twelve FinalTest144 cells, by 3.4% to 14.2% in Mean error and by 2.4% to 14.1% in Final error.
Part 3 lists all rows of the comparison groups. The revision will include these tables and the comparator identities in the appendix, and every table caption will state the evaluation set, the number of seeds and the configuration identity.
Part 3 of 5. Question 2, continued: all comparison groups on FinalTest144
This part lists the rows of every comparison group on FinalTest144. Every cell gives the Mean error and then the Final error, with two seeds per row. A cell that reads "n/a" means that the row is not defined for that task, as in Tables 4 and 5 of the paper.
The first table gives the TEMPO entry and the same-backbone controls. On the five main tasks the TEMPO entry is below every control of the original pool, which are [r38] and the two fine-tuning columns. [r38] is the TEMPO variant that applies the projection per parameter tensor. Direct fine-tuning is the row [r20], and for CE-RPUI the row [r36]. Recurrent fine-tuning is the row [r21], and for NS-SL and NS-SVS the row [r51]. The last column is the appended baseline portfolio [r13] described in part 2. For ACE the TEMPO entry itself is direct fine-tuning, as part 1 explains.
| Task | TEMPO entry | [r38] | Direct fine-tuning | Recurrent fine-tuning | [r13], appended |
|---|---|---|---|---|---|
| NS-PwC | 9.121 / 12.385 | 9.255 / 12.723 | 14.322 / 23.398 | 17.608 / 27.755 | 9.378 / 12.689 |
| NS-SL | 17.373 / 28.083 | n/a | 29.642 / 59.392 | 17.947 / 28.761 | 18.546 / 30.106 |
| NS-SVS | 20.181 / 34.668 | n/a | 30.518 / 67.908 | 20.850 / 35.921 | 22.456 / 38.263 |
| FNS-KF | 9.573 / 13.638 | 9.890 / 14.242 | 17.216 / 27.895 | 20.615 / 33.074 | 9.806 / 13.995 |
| CE-RPUI | 136.024 / 206.903 | n/a | 137.598 / 208.858 | 152.613 / 235.021 | 134.090 / 197.861 |
| ACE | 23.953 / 42.498 | n/a | n/a | 31.783 / 53.840 | n/a |
The second table gives the three DPOT-L configurations.
| Task | Native fine-tuning | Score-H4 | TEMPO-AR |
|---|---|---|---|
| NS-PwC | 52.007 / 80.516 | 44.258 / 70.235 | 44.629 / 69.131 |
| NS-SL | 50.893 / 78.682 | 41.320 / 65.276 | 45.277 / 68.666 |
| NS-SVS | 30.368 / 49.154 | 26.419 / 45.222 | 29.334 / 47.346 |
| FNS-KF | 52.064 / 80.387 | 44.077 / 69.572 | 44.742 / 69.107 |
| CE-RPUI | 314.159 / 400.839 | 272.004 / 355.193 | 280.004 / 357.860 |
| ACE | 13.961 / 21.391 | 13.598 / 22.176 | 13.146 / 20.879 |
On the five main tasks every external architecture is above TEMPO in both metrics. The third table gives the 13 external architectures, and its last row is the strongest ACE comparator of part 2.
| External architecture | NS-PwC | NS-SL | NS-SVS | FNS-KF | CE-RPUI | ACE |
|---|---|---|---|---|---|---|
| FNO | 70.569 / 115.659 | 82.723 / 142.969 | 63.217 / 117.499 | 73.514 / 121.103 | 300.237 / 406.114 | 70.557 / 107.090 |
| F-FNO | 97.769 / 161.752 | 120.152 / 193.866 | 102.833 / 178.416 | 97.794 / 161.443 | 363.259 / 482.087 | 71.308 / 114.007 |
| U-NO | 227.084 / 366.666 | 155.669 / 238.432 | 80.078 / 141.623 | 238.489 / 377.606 | 676.307 / 793.335 | 283.536 / 396.394 |
| LSM | 261.878 / 420.592 | 137.856 / 222.493 | 53.791 / 96.419 | 254.680 / 409.117 | 629.200 / 767.757 | 134.281 / 203.996 |
| FactFormer | 484.147 / 711.294 | 424.772 / 635.899 | 320.859 / 483.947 | 481.645 / 705.819 | 598.883 / 651.489 | 39.365 / 61.616 |
| Transolver | 479.129 / 741.561 | 440.147 / 664.103 | 214.870 / 378.085 | 464.018 / 741.148 | 610.784 / 660.493 | 73.084 / 109.511 |
| UPT | 493.847 / 730.336 | 456.948 / 670.325 | 556.155 / 740.537 | 492.956 / 727.289 | 639.906 / 684.111 | 19.851 / 34.680 |
| DRIFT-Net | 567.465 / 715.621 | 518.629 / 633.049 | 461.515 / 575.420 | 579.053 / 721.921 | 658.596 / 710.066 | 329.909 / 397.549 |
| SGNO | 204.414 / 332.049 | 257.379 / 409.923 | 224.690 / 387.989 | 220.866 / 351.835 | 500.024 / 608.705 | 58.680 / 105.194 |
| Flower-Tiny | 38.677 / 64.343 | 63.425 / 101.688 | 40.592 / 71.730 | 53.840 / 89.488 | 283.252 / 391.779 | 10.159 / 17.510 |
| DPOT-M | 52.517 / 81.320 | 48.024 / 75.262 | 31.645 / 50.896 | 52.970 / 82.308 | 282.057 / 367.565 | 16.079 / 24.563 |
| CNO-FM | 160.917 / 259.258 | 139.781 / 219.462 | 74.052 / 129.108 | 150.296 / 241.250 | 542.671 / 700.652 | 112.047 / 170.832 |
| PDE-Transformer | 192.264 / 323.264 | 251.062 / 401.525 | 261.177 / 422.975 | 209.680 / 357.267 | 465.603 / 606.957 | 61.589 / 101.599 |
| Flower-Tiny expert with residual head [r52] | n/a | n/a | n/a | n/a | n/a | 8.132 / 13.216 |
Part 1 of 5. Question 1 and Weakness 1: the numbers in Appendix A and the evaluation scope
Thank you for the careful reading of the appendices. We answer Question 1 and Weakness 1 here, Question 2 in parts 2 and 3, and Question 3 with Weakness 2 in parts 4 and 5.
Q1. All three sets of values in Appendix A belong to one task, Allen-Cahn, which we abbreviate as ACE. ACE is the sixth task of the original registry, and its entry in the TEMPO-PDE row is direct fine-tuning. Tables 4 and 5 display the other five of the six registered tasks, and the revision will show all six. The table below identifies each set of values. All error values in our five parts are multiplied by 1000.
With Poseidon-B, the direct TEMPO recipe lowers the error on ACE. We trained the submitted direct TEMPO recipe on ACE with two seeds and the same 1,280 updates. The recipe lowers the error of the fine-tuning entry in both seeds and on every trajectory, and this holds on all three evaluation sets of the table below. The table gives two-seed averages of the Mean and Final error. Its last row is the test set HeldOut128, which we have added. There the two direct TEMPO models are evaluated on the same 128 trajectories as the fine-tuning reference.
In relation to Tables 4 and 5 and Appendix E.3, the three sets of values are the ACE entries of three different rows. The first is the TEMPO-PDE row [r57], the second is the row [r15], and the third is the Flower-Tiny study of Appendix E.3. The four-frame Flowers model and the single-state Flower-Tiny differ in input interface and training budget, so 0.976 and 3.977 belong to two different models. The Flower-Tiny row among the external architectures in Table 1 and in part 3 is the standardized external baseline with its own training runs. It is again a different set of models from the Appendix E.3 study. The revision will rewrite the opening paragraph of Appendix A so that each value is given with its task, model, evaluation set and number of seeds, and it will add the six-task tables to the appendix.
W1. Two test sets support the main comparison independently of Fresh128.
The first is HeldOut128, which is new. It contains 128 trajectories per task that were not used during development, all from the official test files. We fixed the trajectory lists, checkpoints and analysis before inference, and we evaluated 204 models with three seeds for every matched configuration. On this set TEMPO is 5.2% to 42.9% below the stronger fine-tuning reference of each task in Mean error and 6.8% to 49.2% below it in Final error. It is also below all seven alternative training rules in all 42 recurrent comparisons of the general response.
The second is FinalTest144, whose complete results are in parts 2 and 3. On this one-shot test set TEMPO has the lowest Mean and Final error on all five main tasks among all methods of the comparison pool that was frozen before the evaluation.
In addition, results obtained on development data carry over to unused test data. All 72 comparisons of the new three-seed study that were registered on the development set Gate96 keep their sign when the same models are evaluated on HeldOut128.