- Rating
- 2
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper proposes TSComp, a time-series foundation model that explicitly decomposes signals into wavelet scales, models temporally aligned cross-scale interactions, and pre-trains with directional masking plus energy, time-domain, and coefficient reconstruction objectives. The learned representations are transferred through task-specific scale-aware pipelines for forecasting, imputation, classification, and anomaly detection. Results across a large pre-training corpus and multiple downstream benchmarks are competitive or state of the art on several metrics.
Critical strengths
-
The paper identifies a concrete limitation of time-domain reconstruction and connects each architectural component to scale separation, temporal alignment, or cross-scale composition.
-
TSComp is tested on four distinct task families, including 27 classification datasets and 245 anomaly datasets, rather than being optimized for forecasting alone.
-
The scale-wise reconstruction study, cross-scale correspondence probe, component ablations, and scale-aware adaptation comparison provide evidence beyond aggregate task scores.
Critical weaknesses
-
The method uses Symlet-4 and a fixed scale hierarchy throughout. The paper acknowledges this limitation, but does not provide enough sensitivity analysis across wavelet families, decomposition depths, nonstationary settings, or datasets with irregular sampling.
-
TSComp changes the pre-training representation, backbone, objectives, and downstream heads at the same time. Since the proposed scale-aware task pipelines themselves improve performance, the comparison does not establish how much of the gain comes from cross-scale pre-training rather than task-specific architectural adaptation. Compute, parameter, and pre-training-data matching against all foundation-model baselines is also not fully clear.
-
Removing coefficient reconstruction or scale-energy calibration improves forecasting in the reported ablation, TSComp-base is worse than TSComp-small for classification, and GTM remains better on two anomaly metrics. The near-perfect reconstruction and correspondence-probe scores also deserve scrutiny because they may partly reflect an easy diagnostic aligned with TSComp's training objective.
Questions for authors
-
Can the authors provide a controlled comparison using the same backbone and downstream head, varying only the pre-training objective, and report matched parameter counts and pre-training compute for the main baselines?
-
How sensitive are the results to wavelet family, decomposition depth, patch alignment, and the magnitude or type of cross-scale misalignment used by the probe?
6 comments
Patch alignment: Time-aligned patching preserves temporal correspondence across wavelet scales, enabling SCA to model interactions among scale-specific structures describing the same temporal region and MDM to encourage reconstruction from complementary cross-scale context. We compare TSComp with TSComp-NoAlign, which flattens coefficient patches into a single token sequence and applies conventional self-attention without explicitly enforcing temporal alignment in cross-scale interactions. As reported in Table 1, TSComp achieves lower MSE across all four evaluated settings, supporting the effectiveness of explicitly modeling cross-scale relationships among aligned coefficient representations
Table 1. Comparison of TSComp with TSComp-NoAlign (MSE).
| Model | Forecasting ETTm1 | Forecasting Weather | Imputation ETTh1 | Imputation ETTm1 |
|---|---|---|---|---|
| TSComp | 0.3609 | 0.2340 | 0.1813 | 0.0849 |
| TSComp-NoAlign | 0.3651 | 0.2341 | 0.1920 | 0.0854 |
Robustness to Non-stationary Conditions: We investigate TSComp's forecasting performance under non-stationary perturbations, including a comparison with TSComp-Fourier, a variant that replaces wavelet decomposition with Fourier decomposition. We conduct 96-step forecasting on the Exchange dataset, selected because its exchange-rate series exhibit inherently non-stationary temporal dynamics. We introduce three representative perturbations: Trend, which introduces gradual temporal drift; Level Shift, which introduces abrupt changes in the signal mean; and Frequency Regime, which introduces changes in oscillatory frequencies across temporal segments. Each perturbation is applied at three severity levels (L1–L3) to the raw time series before standardization, with normalization statistics computed exclusively from the perturbed training split.
Across all nine evaluated conditions in Table 2, TSComp-small achieves the lowest forecasting MSE, consistently outperforming TSComp-Fourier and the other evaluated baselines. Wavelet decomposition does not require a stationarity assumption and provides temporally localized, scale-specific representations, enabling TSComp to model cross-scale relationships through SCA as temporal dynamics change. In particular, the improvements over TSComp-Fourier suggest that temporally localized wavelet representations are beneficial for cross-scale compositional learning under non-stationary conditions. These findings support TSComp's robustness to the evaluated changes in temporal dynamics.
Table 2. Forecasting performance (MSE) under three types of non-stationary perturbations at three severity levels (L1–L3) on the Exchange dataset.
| Condition | Level | TSComp | TSComp-Fourier | MOMENT | TSPulse |
|---|---|---|---|---|---|
| Trend | L1 | 0.102 | 0.135 | 0.155 | 0.144 |
| L2 | 0.068 | 0.091 | 0.108 | 0.097 | |
| L3 | 0.025 | 0.033 | 0.040 | 0.036 | |
| Level Shift | L1 | 0.136 | 0.169 | 0.206 | 0.172 |
| L2 | 0.158 | 0.198 | 0.251 | 0.200 | |
| L3 | 0.171 | 0.218 | 0.270 | 0.221 | |
| Frequency Regime | L1 | 0.153 | 0.180 | 0.194 | 0.216 |
| L2 | 0.229 | 0.260 | 0.270 | 0.328 | |
| L3 | 0.329 | 0.372 | 0.378 | 0.470 |
Robustness to Irregular Sampling: We investigate TSComp's robustness to irregular sampling using six forecasting benchmarks (ETTh1, ETTh2, ETTm1, ETTm2, Weather, and Exchange), selected for a computationally efficient sensitivity analysis. We use a 512-step lookback window and four prediction horizons (96, 192, 336, and 720). To simulate irregular sampling, we randomly remove observations at missing ratios of 10%, 30%, and 50%, with all channels sharing the same missing timestamps. Missing values are replaced with zeros, while forecasting targets remain fully observed. We report the average MSE across the six datasets and four prediction horizons.
As shown in the Table 3, TSComp-small achieves the lowest average MSE under fully observed conditions and at missing ratios of 10% and 30%, outperforming all compared baselines. At 50% missingness, FreTS achieves better performance, while TSComp-small continues to outperform SimpleTM and TimeMixer. These results suggest that TSComp's multi-scale representations remain effective under moderate observation loss, although performance deteriorates when observations become substantially sparse. Since wavelet decomposition operates on a regular time grid, extensive missingness may distort scale-specific temporal structures and their cross-scale relationships, limiting the effectiveness of the proposed framework under severe observation sparsity.
Table 3. Forecasting performance (MSE) under simulated irregular sampling.
| Model | 0% | 10% | 30% | 50% |
|---|---|---|---|---|
| TSComp-small | 0.347 | 0.344 | 0.432 | 0.644 |
| FreTS | 0.434 | 0.442 | 0.461 | 0.474 |
| SimpleTM | 0.408 | 0.428 | 0.567 | 0.812 |
| TimeMixer | 0.356 | 0.412 | 0.535 | 0.776 |
Effect of Cross-scale Pre-training
(W2-1) the comparison does not establish how much of the gain comes from cross-scale pre-training rather than task-specific architectural adaptation.
To examine the contribution of cross-scale pre-training independently of the proposed scale-aware downstream adaptation, we compare TSComp without scale-aware adaptation (w/o Scale) against a capacity-matched MOMENT model pre-trained on the same corpus using conventional time-domain masked reconstruction. By excluding TSComp's task-specific scale-aware adaptation and controlling for pre-training data and model size, this comparison provides a more direct assessment of the benefits of cross-scale compositional pre-training over conventional time-domain reconstruction-based pre-training. As shown in the table 4, TSComp w/o Scale consistently outperforms the capacity-matched MOMENT across all three evaluated tasks. These improvements are achieved without the proposed scale-aware downstream adaptation, supporting the contribution of TSComp's cross-scale compositional pre-training framework to transferable representation learning.
Table 4. Comparison of cross-scale compositional pre-training with a capacity-matched time-domain pre-training baseline
| Variant | ETTh1 Imp. MSE | Classification Avg. Acc. | Anomaly VUSPR |
|---|---|---|---|
| Capacity-matched MOMENT | 0.3024 | 47.67 | 0.0266 |
| TSComp w/o Scale | 0.1885 | 67.00 | 0.0343 |
Effect of Pre-training Objectives
(Q1-1) Can the authors provide a controlled comparison using the same backbone and downstream head, varying only the pre-training objective,
Table 3(c) compares variants trained with different combinations of pre-training objectives while keeping the backbone architecture and downstream adaptation settings unchanged. The results show that the effects of individual objectives vary across downstream tasks, with some ablated variants outperforming the complete configuration in specific settings.
TSComp's pre-training objectives are designed to jointly capture scale-specific temporal structures and their cross-scale compositional relationships within a unified framework, rather than to maximize performance on individual downstream tasks. To assess whether the proposed pre-training objectives achieve these intended representational properties, we conduct additional diagnostic ablations using the same backbone and probing protocol (Table 5). The complete configuration achieves the strongest combination of scale-specific reconstruction fidelity and cross-scale correspondence discriminability. While using only preserves high scale-wise reconstruction fidelity, it yields lower correspondence probe performance. Conversely, using only leads to weaker fine-scale reconstruction (D1 correlation drops from 0.999 to 0.665). These findings suggest that the objectives provide complementary learning signals for capturing scale-specific temporal structures and their cross-scale relationships, although their contributions to downstream performance remain task-dependent.
Table 5. Diagnostic analysis of pre-training objective ablations. Reconstruction fidelity is measured by Pearson correlation in the time domain and at individual wavelet scales.
| Variant | Time | A3 | D3 | D2 | D1 | B.Acc | AUROC | F1 |
|---|---|---|---|---|---|---|---|---|
| TSComp | 0.999 | 1.000 | 1.000 | 0.999 | 0.999 | 0.999 | 1.000 | 0.999 |
| w/o | 0.995 | 0.994 | 0.998 | 0.998 | 0.998 | 0.990 | 1.000 | 0.990 |
| w/o | 0.999 | 0.999 | 0.998 | 0.996 | 0.996 | 0.991 | 0.999 | 0.991 |
| w/o | 0.999 | 1.000 | 1.000 | 0.999 | 0.999 | 0.981 | 0.998 | 0.981 |
| Only | 0.997 | 0.998 | 0.995 | 0.993 | 0.996 | 0.919 | 0.962 | 0.920 |
| Only | 0.927 | 0.950 | 0.873 | 0.813 | 0.665 | 0.986 | 0.999 | 0.985 |
Dear Reviewer cTe8,
We thank the reviewer for their careful evaluation and constructive feedback. In response to their comments, we have (1) clarified our wavelet selection rationale and examined sensitivity to decomposition choices and patch alignment (Table 1; Appendix E), (2) conducted additional robustness evaluations under non-stationary perturbations and simulated irregular sampling (Tables 2–3), (3) introduced a capacity-matched MOMENT baseline and additional pre-training objective ablations to examine the contribution of cross-scale compositional learning (Tables 4–6), (4) strengthened the diagnostic evaluation with additional probing controls and misalignment sensitivity analyses (Tables 7–8), and (5) clarified the interpretation of task-dependent ablation results and model scaling behavior. We will incorporate these additional experiments and clarifications into the revised manuscript. Detailed responses are provided below.
Sensitivity on Wavelet Decomposition and Analysis on Challenging Settings
We appreciate the reviewer for highlighting the importance of sensitivity analysis. We emphasize that the primary contribution of TSComp lies in its pre-training framework for explicitly learning scale-specific temporal structures and their cross-scale compositional relationships, not in selecting a particular wavelet basis.
We adopt Symlet-4 with a three-level decomposition as the default configuration based on its structural compatibility with our pre-training design. Specifically, Symlet-4 combines orthonormality, compact support, near symmetry, and four vanishing moments. Orthonormality ensures energy preservation and supports our theoretical formulation of scale-wise reconstruction and scale-energy distribution calibration, while compact support provides temporal localization for modeling scale-specific structures and their cross-scale correspondence. Thus, our choice of Symlet-4 is motivated by its suitability for the proposed learning objectives, without assuming that it achieves optimal performance on every downstream task.
We investigate the sensitivity of TSComp to wavelet bases (Appendix E.3, Table 23), decomposition depths (Appendix E.2, Figure 6), and alternative decomposition strategies (Appendix E.1, Table 22).
Wavelet basis (Appendix E.3): Table 23 compares Symlet-4, Bior 4.4, Db4, and Haar across four downstream tasks. Symlet-4 achieves the best forecasting performance on ETTh1, whereas Bior 4.4, Db4, and Haar perform best in imputation, classification, and anomaly detection, respectively. These results indicate that no single wavelet basis is optimal across all tasks. As discussed above, our choice of Symlet-4 is motivated by its structural compatibility with cross-scale compositional pre-training, not by task-specific performance optimization.
Decomposition depth (Appendix E.2): Figure 6 evaluates decomposition levels from 1 to 3. The results show that the three-level decomposition achieves the lowest forecasting and imputation MSE on ETTh1, supporting the benefit of representing temporal structures at multiple resolutions.
Decomposition strategy (Appendix E.1): Table 22 compares orthonormal DWT with Fourier decomposition, learnable wavelets, and residual pyramids. Orthonormal DWT achieves the best forecasting and imputation performance on ETTh1, while learnable wavelets yield slightly higher classification accuracy. Although orthonormal DWT is not optimal for every task, our choice prioritizes the four structural requirements for cross-scale compositional pre-training—temporal localization, an explicit scale hierarchy, perfect recomposition, and energy preservation—while maintaining competitive performance across downstream tasks.
In response to the reviewer's comment, we have extended our sensitivity analysis to include patch alignment, nonstationary settings, and irregularly sampled datasets.