- Rating
- 2
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
HarMoTC is a frequency-domain framework for multivariate long-term time-series forecasting. On top of a shared spectral backbone (instance normalization → RFFT → adaptive spectral gating → linear cross-variable mixing Q∈R^{C×C} → IRFFT → temporal projection), it adds two components: (1) calendar-conditioned spectral modulation, which encodes the forecast-origin hour-of-day/day-of-week indices into K Fourier harmonics and maps them through an MLP (a FiLM-style generator, per Table 8) to channel- and frequency-specific parameters (γ,β) applied in asymmetric pre- and post-modulation steps (Eqs. 8–9); and (2) phase–amplitude decoupled spectral modeling, which polar-decomposes the conditioned spectrum and applies learnable frequency-wise amplitude gains g and phase shifts δ generated from two temporal kernels (Eqs. 11–12). Training uses a pure complex-spectral MAE objective (Eq. 14). Validation covers 7 benchmarks × 4 horizons against 9 baselines, with core-design ablations, calendar-stratified spectral fidelity, a counterfactual calendar-alignment intervention with bootstrap CIs, efficiency, and a three-seed robustness study. Headline claims: best or tied-best on 11/14 dataset–metric pairs; macro-averaged MSE/MAE reductions of 3.18%/1.64% over the strongest baseline (SSformer).
Critical strengths
A causal-style evidence chain for the central mechanism, rare in this genre. The counterfactual intervention (Tables 4/9) demonstrates — per horizon, with CIs — that the model functionally relies on correctly aligned calendar semantics, and the calendar-stratified spectral analysis (Table 3, 96/96 phase-group wins) independently corroborates the channel through which conditioning acts. This is consequential because it elevates the paper from "ablation shows a drop" to genuine mechanism-level evidence, and it is the strongest reason the work deserves serious consideration.
Critical weaknesses
The second headline contribution — phase–amplitude decoupled modeling — is weakly evidenced and arbitrarily designed. The ablation shows the two spectral branches jointly contribute markedly less than calendar conditioning (w/o Amp&Phase degrades 0.2785→0.2908 MSE, +4.4%; w/o TC degrades to 0.2968, +6.55%), and the conditioning path itself barely touches phase: pre-modulation (Eq. 8) rescales only the real part and divides by (1+|γ|+ε), post-modulation (Eq. 9) multiplies by (1+γ) — neither symmetric nor inverse — and Appendix A.3 concedes the mechanism "is not a fully general complex-valued transformation." This is consequential because the title, abstract, and contribution list place phase modeling on equal footing with temporal conditioning; the evidence instead indicates the effective contribution is calendar FiLM, with the phase–amplitude machinery adding a modest increment whose design is never justified.
Questions for authors
1: What justifies the asymmetry between Eq. 8 (real-part only, divide by 1+|γ|+ε) and Eq. 9 (multiply by 1+γ, add β)? Did you try full complex-valued modulation of real/imaginary (or amplitude/phase) parts, and what were the results?
2: The ablation omits Traffic. Does removing temporal conditioning improve Traffic (i.e., is calendar conditioning harmful for heterogeneous sensor networks)? This materially affects the interpretation of the Traffic failure.
6 comments
Response to Reviewer CSiB (2/2): Alternative modulations and Traffic ablation
Table numbers refer to the revised manuscript.
Q1. Results of alternative modulation designs
For and , we evaluated:
- Symmetric real: ; pre-modulation , post-modulation .
- Full complex affine: , ; pre-modulation , post-modulation . Two additional heads generate the imaginary gain and bias.
- Direct calendar amplitude-phase: ; pre-modulation , post-modulation , with log-gain and phase rotation .
Both complex variants preserve the real-valued constraints at DC and Nyquist frequencies. All three retain the original amplitude and phase branches, so this comparison tests calendar injection. These results are included in revised Table 3.
Results use and seeds {2024, 2025, 2026}. Entries are mean ± sample SD; means and parameters are truncated to three decimals, SDs to four. Parameters are in millions (M).
| Dataset | Metric | Original | Symmetric real | Complex affine | Calendar amp-phase |
|---|---|---|---|---|---|
| ETTh1 | MSE | 0.353±0.0011 | 0.360±0.0077 | 0.364±0.0047 | 0.371±0.0034 |
| MAE | 0.385±0.0013 | 0.387±0.0043 | 0.391±0.0015 | 0.396±0.0020 | |
| Params (M) | 0.906 | 0.906 | 1.083 | 0.906 | |
| Weather | MSE | 0.147±0.0008 | 0.149±0.0007 | 0.150±0.0006 | 0.158±0.0005 |
| MAE | 0.189±0.0011 | 0.190±0.0010 | 0.192±0.0009 | 0.202±0.0006 | |
| Params (M) | 0.801 | 0.801 | 0.934 | 0.801 | |
| Electricity | MSE | 0.140±0.0001 | 0.142±0.0004 | 0.151±0.0003 | 0.161±0.0004 |
| MAE | 0.239±0.0002 | 0.241±0.0004 | 0.251±0.0008 | 0.261±0.0005 | |
| Params (M) | 2.395 | 2.395 | 3.433 | 2.395 | |
| Traffic | MSE | 0.435±0.0009 | 0.438±0.0019 | 0.434±0.0009 | 0.468±0.0025 |
| MAE | 0.287±0.0011 | 0.291±0.0029 | 0.288±0.0002 | 0.314±0.0009 | |
| Params (M) | 4.784 | 4.784 | 7.572 | 4.784 |
The original has the lowest mean MSE and MAE on ETTh1, Weather, and Electricity. On Electricity, complex affine and direct calendar amplitude-phase modulation increase mean MSE by 7.39% and 15.00%.
On Traffic, complex affine has 0.19% lower mean MSE, but 0.32% higher mean MAE and 58.26% more parameters; the original has the lowest mean MAE. Percentages use untruncated values. These results support the original modulation's accuracy/parameter trade-off in the reported settings; additional complex modulation freedom does not consistently improve both metrics.
Q2. Does removing temporal conditioning improve Traffic?
No. We evaluated w/o TC at four horizons with and seeds {2024, 2025, 2026}, removing both calendar-conditioned pre- and post-modulation while retaining the other forecasting components.
Entries are MSE / MAE. The full-model column is the single-run reference from Table 13; the remaining columns are additional w/o TC runs.
| Horizon | HarMoTC (Table 13) | w/o TC, 2024 | w/o TC, 2025 | w/o TC, 2026 |
|---|---|---|---|---|
| 96 | 0.435 / 0.287 | 0.468 / 0.314 | 0.466 / 0.311 | 0.466 / 0.311 |
| 192 | 0.452 / 0.298 | 0.482 / 0.319 | 0.479 / 0.318 | 0.481 / 0.319 |
| 336 | 0.473 / 0.314 | 0.497 / 0.326 | 0.494 / 0.324 | 0.497 / 0.326 |
| 720 | 0.505 / 0.336 | 0.555 / 0.359 | 0.551 / 0.357 | 0.548 / 0.354 |
Averaging horizons within each seed, then computing mean ± sample SD across seeds from the displayed values, gives approximately MSE 0.4986 ± 0.0016 and MAE 0.3281 ± 0.0011 (truncated to four decimals).
For the same four-horizon summary, Table 6 reports the full model's three-seed results as MSE 0.466 ± 0.003 and MAE 0.309 ± 0.003. Removing conditioning therefore increases average MSE/MAE by approximately 7.0%/6.2%. Every w/o TC entry also exceeds the corresponding Table 13 full-model reference.
Calendar conditioning thus remains beneficial on Traffic. Removing it does not resolve the gap to strong baselines, so that gap cannot simply be attributed to harmful calendar conditioning in heterogeneous sensor networks.
Thank you again for your constructive feedback. We would be happy to provide further clarification on any remaining concerns.
Dear Reviewer CSiB,
On Weather (L = T = 96), we further examine phase calibration through forecasting results and an output-phase diagnostic from repeated runs. Both variants retain calendar conditioning and amplitude rescaling.
| Metric | Full | w/o Phase | Reduction |
|---|---|---|---|
| MSE | 0.146 | 0.149 | 2.01% |
| MAE | 0.187 | 0.190 | 1.58% |
| Weighted phase error | 0.602264 | 0.625555 | 3.72% |
MSE/MAE are truncated to three decimals, and their reductions use the displayed values. The phase diagnostic (seed 2024; all 10,444 test windows) shows a 3.72% reduction, with a 95% paired block-bootstrap interval of [0.008178, 0.036860] for w/o Phase − Full. This complements our ablations by linking phase calibration to improved output phase alignment within the calendar-conditioned framework.
We are writing to ask whether our rebuttal has addressed your concerns. We would be happy to discuss any remaining questions.
Clarifying HarMoTC's Technical Contributions, Modulation Design, and Traffic Ablation
Response to Reviewer CSiB (1/2): Contribution positioning and design rationale
Thank you for recognizing the calendar-alignment intervention and spectral analysis. We clarify the contribution hierarchy and design below; Part 2 provides the modulation comparisons and Traffic ablation. Table numbers refer to the revised manuscript.
Contribution and supporting evidence
Calendar-conditioned modulation is the core mechanism for adapting spectral transformations to temporal context. It generates variable- and frequency-specific parameters from calendar indices. Within this framework, phase-amplitude decoupled modeling provides explicit amplitude and phase controls. Both designs can affect amplitude and phase, but their parameter sources and roles differ.
For Z(q)=A(q)exp[iΦ(q)], Eq. (12) gives
U(q)=[A(q)⊙g]⊙exp{i[Φ(q)+δ]}.
Separate temporal kernels generate the frequency-dependent gain g and phase shift δ, shared across samples and variables. Calendar dependence is supplied by the modulation.
Table 2 measures the module's additional benefit with calendar conditioning retained. Removing both branches increases macro-average MSE from 0.2785 to 0.2908 (+4.4%) across four datasets and four horizons. Removing amplitude rescaling alone gives 0.2870; removing phase calibration gives 0.2843 (+2.08%). The latter retains the input phase, calendar modulation, and complex spectral loss. The phase kernel has only 96 parameters at L=96.
Revised abstract, introduction, and contributions
The revised contribution list first presents the integrated framework, then details its core calendar-modulation mechanism. Amplitude-phase modeling remains part of the framework contribution, with its additional benefit evaluated separately. The title describes the complete architecture; the model and equations are unchanged.
Abstract, method description:
Introduction overview (Section 1):
Contributions 1 and 2:
Q1. Why use asymmetric modulation?
Eq. (8) subtracts β and divides by 1+∣γ∣+ϵ. The divisor is bounded below by 1+ϵ, avoiding division by a near-zero value. Eq. (9) applies (1+γ)⊙Rv+β after spectral transformation and cross-variable interaction, allowing signed affine adjustment. These operations serve input scaling and output adjustment, respectively, and are not intended as inverse transformations.
Keeping the imaginary part unchanged does not fix the phase: scaling or shifting the real part can change the argument of the complex coefficient. The calendar-stratified analysis (Table 4) reports 5.18% lower weighted phase error, with improvements in all 96 hour-horizon groups. This evaluates calendar conditioning, while Table 2 separately measures the benefit of learnable phase calibration. Part 2 compares the requested alternatives.