- Rating
- 3
- Confidence
- 3(+1)
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 3(+1)
Read review
Paper summary
This paper introduces STFO for continual spatial-temporal forecasting under expanding sensor networks. By mapping irregular sensor observations onto a fixed latent grid, the authors summarize the spatial-frequency characteristics of the current field to capture the current spatial regime. The method is is evaluated on three datasets across multiple horizons and hidden widths, demonstrating improvements in most settings.
Critical strengths
- The paper claims a meaning and interesting insights in continual spatial-temporal forecasting.
- The SPECTRAL REGIME ENCODER may provide a useful mechanism on changing spatial patterns.
- The experiment results are competitive.
- Theoretical consistency analysis to support their motivation.
Critical weaknesses
- The novelty might be limited.
- The ablation study shows slight performance differences after removing individual component. Since the final prediction combines field and local-history contributions, it remains unclear how much of the overall improvements is attributable to the proposed field modeling components and what truly contributes to the improvements.
Questions for authors
The improvements appear considerably better on traffic forecasting. Could the authors discuss why this may be the case and whether this indicates sensitivity to the dataset domains.
4 comments
Response to Reviewer R3n9
q1. Performance across domains
1. Additional non-traffic datasets
To assess performance beyond traffic, we added temperature forecasting on USRCRN (2009–2014, six periods) and soil-moisture forecasting on CTP-SMTMN (2010–2016, seven periods), using observed station changes. These supplementary experiments compare STFO-L with three selected baselines: STRAP, DOL and FNO. We report three-seed mean ± sample standard deviation after equally averaging over all periods and 12 prediction horizons, evaluated on all current stations. The units are °C for USRCRN and for CTP-SMTMN.
| Dataset | Method | MAE ↓ | RMSE ↓ |
|---|---|---|---|
| USRCRN | STFO-L | 2.64 ± 0.03 | 3.45 ± 0.04 |
| USRCRN | STRAP | 3.26 ± 0.16 | 4.25 ± 0.18 |
| USRCRN | DOL | 2.72 ± 0.04 | 3.54 ± 0.06 |
| USRCRN | FNO | 2.87 ± 0.15 | 3.70 ± 0.15 |
| CTP-SMTMN | STFO-L | 6.00 ± 0.13 | 10.13 ± 0.20 |
| CTP-SMTMN | STRAP | 8.07 ± 1.47 | 13.40 ± 1.66 |
| CTP-SMTMN | DOL | 6.60 ± 0.10 | 11.63 ± 0.27 |
| CTP-SMTMN | FNO | 6.88 ± 0.45 | 11.78 ± 0.50 |
STFO-L outperforms all compared baselines on both additional datasets, with substantial gains on CTP-SMTMN (9.09% lower MAE and 12.90% lower RMSE than the strongest baseline), demonstrating strong performance beyond traffic forecasting.
We have added these supplementary experiments in Appendix L (Table 15), with corresponding descriptions in Sections 4.1 and 4.2.
2. Understanding the smaller gains on AIR-Stream
To examine possible factors behind the smaller gains on AIR, we compare sensor growth and mean Pearson correlations over fixed neighboring sensor pairs from the first-period geographic 8-nearest-neighbor graph, restricted to sensors present in both the first and last periods. Correlations are computed using each period’s training observations and averaged over pairs with valid correlations in both periods. Sensor counts and correlations below are shown as first → last; growth is the net increase in sensor count.
| Dataset | Sensors (first → last) | Growth (%) | Neighbor corr. (first → last) |
|---|---|---|---|
| PEMS-Stream | 655 → 871 | 33.0 | 0.83 → 0.82 |
| CA-Stream | 480 → 1698 | 253.8 | 0.88 → 0.86 |
| AIR-Stream | 1087 → 1202 | 10.6 | 0.68 → 0.74 |
| USRCRN | 15 → 72 | 380.0 | 0.86 → 0.76 |
| CTP-SMTMN | 37 → 57 | 54.1 | 0.77 → 0.92 |
AIR’s characteristics may make the challenges addressed by our method less pronounced. Its sensor count increases by only 10.6%, the smallest growth among the five datasets, suggesting less demand for adaptation to new observation layouts. Its neighboring-sensor correlations are also the lowest in both periods (0.68 → 0.74), which may limit the gains from sharing predictive information across locations. Consequently, decoupling layout adaptation from the underlying forecasting dynamics and reusing predictive structure through a shared field may yield smaller incremental gains on AIR. These observations suggest that the magnitude of the gains may vary with network expansion and spatial dependence across domains.
We have added this analysis in Appendix M (Table 16) and summarized the findings in Section 4.2.
We thank you again for the constructive feedback, which has helped us improve the clarity of the paper and strengthen the supporting evidence. All revisions are highlighted in red in the newly uploaded PDF for your review.
Thank you for the detailed response and for addressing my concerns. I appreciate the authors’ efforts and the additional experiments. I will maintain my original score. I wish the authors all the best with their work.
Thank you for your follow-up and for taking the time to consider our response and additional experiments. We appreciate your thoughtful feedback, which has helped us improve the paper.
Response to Reviewer R3n9
We thank the reviewer for the careful review and constructive suggestions. Below, we clarify the contribution, quantify the roles of the two prediction branches, and discuss performance across domains.
Our core contribution is a mechanism for reusing and adapting spatial forecasting knowledge as sensor networks expand. The key insight is that changes in the observation layout need not entail equally large changes in the underlying spatial dynamics. CFE and CQD provide coordinate-based observation and query interfaces to a shared latent field, allowing the same parameterization to operate across sensor layouts; SRE adjusts the operator response according to the current spectral state. Together with our consistency analysis, these designs form a continual forecasting framework that supports knowledge reuse across layouts and adaptation to changing spatial patterns.
These mechanisms are described in Sections 3.1–3.5, with the consistency analysis provided in Appendix C.
To isolate the two branches, we independently retrained field-only and local-history-only variants over all periods of PEMS-Stream and AIR-Stream. We report three-seed mean ± standard deviation of MAE, after averaging each run over all periods and 12 prediction horizons.
Additionally, field-only outperforms local-history-only on both datasets and approaches Full on PEMS, indicating that field modeling provides the main predictive capability, with local history offering complementary gains. Full reduces MAE relative to local-history-only by 33.80% and 8.82% on PEMS and AIR, respectively, and further improves over field-only by 0.99% and 3.42%.
Removing a single field component leaves the remaining field components active, so the small gaps in the original ablations reflect individual components’ incremental effects rather than the contribution of the entire field branch.
We have added these independently retrained branch ablations and their analysis in Section 4.3 and Figure 3(d).