- Rating
- 3
- Confidence
- 3
- Informativeness
- 3
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
This paper presents TS-CAN, a time series forecasting model built around Clifford geometric products. It uses inner-product and wedge-product terms to model alignment and directional relationships, and combines temporal and channel-wise interactions with multi-scale context. The paper also relates the wedge term to discrete Lévy area and the quadrature spectrum of cross-covariance. TS-CAN reports the best or tied MSE in 28 of 32 long-term forecasting settings and competitive results on short-term tasks.
Critical strengths
-
Novel idea. I find the use of Clifford geometric products for forecasting interesting. The symmetric and antisymmetric terms offer a clear way to think about directional information, and the links to Lévy area and spectral analysis help motivate the design.
-
Strong results. TS-CAN performs well on the tested benchmarks, particularly Traffic and Exchange. The attention free design and reported inference efficiency are also appealing.
Critical weaknesses
- Ablations are limited to one main settingMost component ablations are run on ETTh1-96. I would like to see whether the temporal and channel-wise wedge terms also help on Traffic and Exchange, where the model shows its largest gains. Table 5 also suggests that the benefit of the wedge term varies across ETT datasets. In addition, the unstructured bilinear baseline in Table 3(a) changes the activation as well as the parameterization, so it does not completely isolate the effect of explicit antisymmetry.
2.Incomplete efficiency comparisonThe reported memory use and inference latency look promising. However, TimeMixer++, the strongest accuracy baseline in the main experiments, is missing from Tables 6 and 9. Including it, preferably using the configurations behind the accuracy results, would make the claimed accuracy-efficiency trade-off easier to assess.
Questions for authors
-
Could the authors report separate temporal- and channel-wise wedge ablations on Traffic and Exchange, preferably for a longer prediction horizon as well?
-
In Table 3(a), does the explicit antisymmetric parameterization still outperform the unstructured bilinear baseline when the SiLU activation and other components are kept the same?
-
Could the authors include TimeMixer++ in the efficiency comparison and report accuracy and latency for the configurations used in the forecasting experiments?
8 comments
Response to Reviewer rbt8 [2/4]
Traffic and Exchange. Both wedge terms help at every horizon on both datasets.
- Traffic. Removing either wedge costs 9.4%–14.3%, and removing both costs 11.5%–15.7%. These are the largest and most stable effects among all datasets. Traffic sensors share daily and weekly cycles with sensor-specific phase offsets, which is the lead–lag structure that the antisymmetric term is designed to capture.
- Exchange. The effect is largest at the longest horizon. At , removing the temporal wedge, the channel wedge, or both increases MSE by 14.5%, 12.4%, and 18.4%.
Complementarity at long horizons. On ETTh1 at , removing only one wedge costs at most 1.3%, but removing both costs 6.7%. A similar pattern holds on ETTh1 at and on Exchange and Traffic at . The two branches therefore carry partly overlapping directional information, and losing both is clearly worse than losing either.
Variation across ETT datasets. The wedge terms help on all four ETT datasets, and the size of the benefit depends on both dataset and horizon.
- ETTh1 and ETTh2. The gain is clear at every horizon, reaching 16.4% on ETTh2 at .
- ETTm2. The gain is moderate at intermediate horizons and larger at (5.0%–7.8%).
- ETTm1. The effect is smallest beyond , where the variants stay within 1.8% of TS-CAN. At the gain is still 6.3%–7.0%.
- ETTh2 in the surrogate study (Table 6). The full model did not reach its main-table accuracy (0.305 vs. 0.280), because that study uses a single training configuration (App. B). Under the main configuration, the wedge terms help on ETTh2 at all four horizons.
This pattern is consistent with the wedge acting as an estimator of directional structure (Props. 1–2). It helps where such structure is present and remains predictable over the forecast window, and on ETTh1 and ETTm1 its gain disappears once IAAFT surrogates remove this structure.
Response to Reviewer rbt8 [3/4]
Q2: Activation-matched antisymmetry comparison
Yes. To answer this directly, we designed controls that keep the SiLU activation, the context, and all other components of TS-CAN fixed, and ran them on all eight datasets at all four horizons. Let and . TS-CAN feeds to the projection . The controls are:
- V1, sign flip: ;
- V2, activation-matched unstructured interaction: ;
- V3, ungated: ;
- V4, symmetric in-phase term: ;
- V5, gated wedge: ;
- V6, strict superset: .
V1–V5 have exactly the same parameters and FLOPs as TS-CAN. V6 has 26% more parameters and can represent TS-CAN exactly. The table reports the average MSE of TS-CAN on each dataset and the relative change of each control.
| Dataset | TS-CAN | V1 | V2 | V3 | V4 | V5 | V6 |
|---|---|---|---|---|---|---|---|
| ETTh1 | 0.412 | +5.9% | +8.9% | +7.5% | +8.8% | +6.5% | +7.1% |
| ETTh2 | 0.341 | +8.7% | +9.6% | +8.7% | +9.1% | +9.5% | +8.2% |
| ETTm1 | 0.352 | +2.2% | +2.2% | +2.7% | +2.6% | +1.7% | +2.3% |
| ETTm2 | 0.258 | +3.8% | +3.8% | +3.4% | +3.1% | +3.3% | +3.7% |
| Electricity | 0.164 | +6.7% | +6.9% | +6.2% | +5.9% | +6.4% | +5.8% |
| Weather | 0.224 | +7.9% | +7.6% | +7.2% | +7.3% | +6.9% | +7.8% |
| Traffic | 0.387 | +12.7% | +13.0% | +12.1% | +11.6% | +12.5% | +11.5% |
| Exchange | 0.338 | +3.5% | +10.2% | +2.1% | +11.6% | +10.9% | +17.9% |
| Average MSE | 0.309 | +6.5% | +8.1% | +6.4% | +7.8% | +7.5% | +8.3% |
| Average MAE | 0.338 | +3.2% | +3.8% | +3.0% | +3.6% | +3.5% | +4.0% |
Across all 32 dataset–horizon settings, TS-CAN has lower MSE than the control in 189 of the 192 comparisons, and the three exceptions are within 0.6%.
Sign-flip control (V1). It is the most direct test of explicit antisymmetry. It keeps the activation, parameters, context, and every other component, and changes only whether the explicit linear term is odd or even. It is worse in 31 of 32 settings, by 6.5% MSE on average, and on ETTh1-96 it is worse on each of three seeds (0.3662 ± 0.0015 vs. 0.3614 ± 0.0014).
Activation-matched control (V2). It applies the same SiLU to both raw products, which is the comparison the reviewer asks for. It is worse in all 32 settings, by 8.1% on average. On ETTh1-96, the setting of the original ablation, it is 2.8% worse, close to the 2.4% gap of the original control. The gap therefore remains once the activation is matched.
Not a capacity effect. The strict superset (V6) is worse in all 32 settings as well. Its largest losses are on Exchange, where the additional freedom overfits a small, non-stationary dataset.
Where the effect is largest. On Traffic, every control is 9.0%–15.4% worse at every horizon. The effect remains clear on Exchange, especially for the controls that remove or gate the explicit wedge (10%–18% on average).
How the two components should be treated. Removing the gate from the in-phase term (V3), symmetrizing it (V4), and moving the gate onto the wedge (V5) each cost 6%–8%, similar to flipping the parity. The best configuration keeps the antisymmetric term explicit, odd, and linear, and applies the nonlinearity only to the in-phase term. This agrees with our analysis: the wedge changes sign under time reversal, whereas SiLU would add an even, reversal-invariant component to it.
Response to Reviewer rbt8 [1/4]
Many thanks to the reviewer for the detailed and insightful comments, which helped us improve the paper. In response, we (i) ran separate temporal and channel wedge ablations on all eight datasets and four horizons, including Traffic and Exchange (W1, Q1). We also (ii) added activation-matched, parameter-matched antisymmetry controls on the same 32 settings (W1, Q2), and (iii) added TimeMixer++ to the efficiency comparison under the same protocol as the other models (W2, Q3).
Q1: Ablations on Traffic and Exchange
We thank the reviewer for this suggestion. We ran the separate temporal and channel wedge ablations on all eight datasets at all four horizons, with the same configuration and training protocol as the main experiments. Each variant removes only the wedge term of the specified branch and keeps the in-phase term and all other components unchanged. The table reports MSE, with the relative change with respect to TS-CAN in parentheses.
TS-CAN has lower MSE than the ablated variant in 92 of the 96 comparisons. There is one tie, and the three exceptions are all within 0.5%. Averaged over the 32 settings, removing the temporal wedge, the channel wedge, or both increases MSE by 7.2%, 7.1%, and 8.1%.