- Rating
- 1
- Confidence
- 4
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 1
Read review
Paper summary
This paper proposes COROS, a score-based diffusion framework for time-series forecasting under corrupted historical inputs. COROS applies diffusion to the concatenated historical and future sequence while retaining the observed history as a condition. It further introduces an imbalanced loss for the historical and future coordinates and optionally uses classifier-free guidance. Experiments are conducted on five probabilistic forecasting datasets and three LTSF datasets with synthetic input corruption.
Critical strengths
- The paper studies a practically relevant problem: historical observations may be corrupted by measurement noise or anomalies, while most conditional forecasting methods assume reliable context.
- The empirical evaluation covers datasets with substantially different dimensionalities, and some robustness improvements.
Critical weaknesses
-
Limited methodological novelty Most components of COROS are already established. Full-sequence diffusion has been explored in methods such as TSDiff; conditional diffusion, VP-SDE, DiffWave-style networks, denoising score matching, and classifier-free guidance are standard techniques. Assigning different weights to historical and future coordinates is also a natural engineering choice. The main contribution is therefore the combination of conditional full-sequence diffusion, latent-history generation, and an imbalanced loss for corrupted-context forecasting. While reasonable, I do not find this combination sufficiently novel without stronger technical or empirical evidence.
-
Presentation and reproducibility The manuscript contains multiple grammatical, notation, algorithm-description, and bibliography issues. Some variables are undefined or unused, several experimental hyperparameters are insufficiently explained, and the reference list contains duplicated or corrupted entries. Some citations also provide only indirect support for the associated claims. These issues are not the primary reason for rejection, but they indicate that the paper requires substantial revision and make the method harder to reproduce.
-
The claimed robustness mechanism is not established. COROS still parameterizes the score as , so the corrupted condition remains directly available to the forecast score at every reverse step. The architecture does not require the forecast to use the supposedly denoised latent history. Moreover, the theoretical factorization relies on , which does not generally follow after independently noising variables with a shared clean-history ancestor. The experiments do not demonstrate that the generated history is actually cleaner than the corrupted input or isolate purification from auxiliary-task regularization.
Questions for authors
1.What prevents the forecast-coordinate score from relying directly on the potentially corrupted condition and ignoring the evolving latent history ? Is this dependence constrained architecturally or only expected to emerge from training?Can the authors quantitatively verify the claimed purification mechanism by measuring whether the final generated history is closer to the clean history than the corrupted input is? Is the generated history quantitatively closer to the clean history than the corrupted input? When , is the historical state still updated during sampling? Under what additional assumptions does the claimed conditional independence hold after forward noising? Does the robustness improvement remain when the forecast score is prevented from directly accessing ?The architecture does not enforce the claimed denoising mechanism。The paper argues that prediction-only diffusion repeatedly uses a corrupted condition during sampling. However, COROS still parameterizes the score as
so the original condition remains directly available to the forecast score at every reverse step. COROS adds the indirect path
but it does not remove the direct path
Nothing prevents the network from ignoring the latent history and continuing to rely directly on the corrupted condition. Moreover, training uses the same clean history as both the condition and the historical generation target. The model is not explicitly trained to recover a clean history from a corrupted condition. The reported gains could therefore arise from auxiliary-task regularization or the enlarged diffusion state rather than actual history purification. The paper does not quantitatively demonstrate that the generated history is cleaner than the corrupted input.
-
Does the robustness improvement remain without classifier-free guidance?
-
In Algorithm 1, the condition is identical to the clean target of the historical diffusion coordinates. Therefore, is a Gaussian centered at , and its optimal score appears to reconstruct exactly. Why should this objective remove a test-time corruption in , rather than reproduce it in the generated history? Is the condition independently corrupted or noised during training in a way not described in Algorithm 1?
5 comments
W3 / Q1. The robustness mechanism.
(a) Direct access to . We agree that the architecture does not enforce reliance on the latent history: the forecast score retains a direct path from . This is by design, since COROS uses the observed context and the evolving latent history jointly, and the existence of a direct path does not by itself determine the overall sensitivity to corruption. What matters is how the trained network uses the two inputs. Appendix B (Table 12) measures this with paired Jacobian diagnostics: at the same diffusion time and latent state, switching the context from clean to corrupted decreases the forecast score's sensitivity to the observed context (median 0.76× on Traffic, 0.66× on Weather) while maintaining or increasing its sensitivity to the latent history (1.02×, 1.12×). On Electricity both increase, the latent-history sensitivity more (1.15× vs. 1.07×).
(b) Is the generated history cleaner than the corrupted input? On real benchmarks the clean history is itself unknown due to its possible inherent noise, so we answer where the ground truth is exact: two ODE systems with spikes injected into 1/8 of the context (Appendix I). With the selected , the generated history largely does not follow the injected spikes while the forecast matches the ground truth (Figure 7). Forecast errors are in Table 17 (2D oscillator: MSE 0.2953 at vs. 0.0006 at ; 3D harmonic: 0.0210 vs. 0.0002; means over five seeds). Regarding auxiliary-task regularization and the enlarged diffusion state: the model in Table 11 has the same enlarged state and the strongest auxiliary history-reconstruction loss, yet it is less accurate and less robust than the selected weight. The gains are therefore not explained by the enlarged state or by more auxiliary history training alone.
(c) Forecasting without direct access to . Removing the direct path would change the conditioning design rather than isolate a component of it. The closest test we ran goes the other way: we replace with a purified estimate (Appendix A.7, following MadSGM), so the forecast never sees the raw corrupted context. For the same model, this increases MSE from 0.34 to 0.89 on clean inputs and from 0.95 to 1.02 at (Table 11). The observed context therefore carries details that the forecast should retain.
(d) . With the history coordinates receive no training signal and COROS reduces to the prediction-only model, which is the prediction-only control reported in Tables 3, 9, 10, and 11.
(e) Conditional independence after forward noising. We agree that does not hold in general; a smooth Gaussian example satisfying our regularity assumptions (A1)–(A3) already violates it, so no such assumption restores it. We no longer rely on it. Lemma 2 now uses the exact decomposition , which requires no independence assumption, and Theorem 1 is re-derived so that it accounts for both the direct dependence on and the dependence through the latent history.
Q2. Does the robustness improvement remain without classifier-free guidance?
Yes. All corrupted-history results in Tables 3, 9, 10, and 11 are sampled without guidance. Guided results are labeled COROS-CFG (Table 2), and every unlabeled COROS entry is unguided; we now state this convention in Section 5.
Q3. If is also the target, why would the model remove rather than reproduce corruption?
We agree at the level of the ideal score on the training distribution: with the optimal history score reconstructs , so the objective alone does not guarantee that corruption is removed. We therefore present the suppression of corrupted context as an empirical property of the trained model rather than a consequence of the objective. Three points qualify the conclusion that the model will reproduce the corruption.
- A corrupted test-time context lies outside the training distribution, so the network's behavior there is not determined by the population-optimal identity. At the forecast level, COROS reduces error growth by 58% relative to the matched prediction-only control on Electricity (Table 10) and has a lower CRPS$_{\mathrm{sum}}$ at every noise level on Electricity and Weather (Table 9).
- The condition is not corrupted during training: training and model selection use clean data, and corruption is applied only at test time (Appendix D; we have made this explicit).
- The toy experiment in Appendix I tests exactly this question. There the training trajectories are noise-augmented and the same augmented sequence is used as both condition and history target, so holds and the model never sees a clean target for a noisy condition. Consistent with the reviewer's analysis, with the generated history reproduces the injected spikes almost exactly (Figure 7, lower panels). With the same architecture suppresses most of them, and the forecast error falls (Table 17). Since a positive does not change the population-optimal score, we attribute this to the finite trained model.
We thank the reviewer again. We believe the corrected theory, the clarified scope, and the new evidence on equal weighting, condition purification, and the toy systems address the main soundness concerns, and we would be grateful if the reviewer would reconsider the assessment in light of these revisions.
We have uploaded the revised manuscript; all changes are highlighted in blue.
We thank the reviewer for the detailed and technically careful review. New results are in Appendix A.7 (Table 11) and Appendix I (Figure 7, Table 17).
W1. Methodological novelty.
We agree that the individual components (VP-SDE, DiffWave-style backbone, DSM, CFG) are standard and that conditional full-sequence diffusion has precedents; we do not claim these components as new. The contribution is (i) the setting: diffusion forecasters trained on clean data and evaluated when the observed context is corrupted at test time; (ii) a model that keeps the observed context c as the condition while jointly evolving a latent history with the forecast, with a history-loss weight γ; and (iii) empirical evidence that this design choice matters.
The new control in Table 11 provides this evidence. With γ=1, the history and forecast losses are weighted equally, which is the full-sequence conditional objective of TSDiff-Cond (Kollovieh et al., 2023), here with our backbone. Under the same corruption protocol, training procedure, and sampler (Traffic MSE):
The selected weight is more accurate at both noise levels and reduces the corruption-induced error growth from 0.61 to 0.53, whereas the equal-weight objective is worse than prediction-only even on clean inputs. (The last row is discussed under Q1(c) below.)
W2. Presentation and reproducibility.
We agree and have revised the manuscript accordingly.