- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 1
Read review
Paper summary
This paper proposes a controllable outlier generation framework based on likelihood reweighting for pretrained score-based diffusion models. Instead of perturbing samples in the high-dimensional data space, the authors formulate outlier generation as reweighting the one-dimensional log-likelihood distribution , and construct the target distribution via exponential tilting. They show that the resulting distribution satisfies first-order stochastic dominance (FOSD) and provide a calibration between the tilting parameter and the Wasserstein-1 outlier strength . At the diffusion level, they show that likelihood reweighting modifies the pretrained score only through a scalar correction along the log-likelihood coordinate; for exponential tilting this correction is available in closed form, , yielding an inference-time controller (Eq. 12) that requires neither retraining nor an auxiliary model. They further provide a manifold convergence analysis showing that the controlled PF-ODE still converges to the data manifold for . Experiments are conducted on nine real-world multivariate time-series datasets, compared against several extreme/outlier-oriented GAN baselines, reporting marginal KS, Tail-MSLE, and additional ACF, correlation error, FGD, and CCDF area metrics, together with an ablation over .
Critical strengths
The empirical study covers nine real-world multivariate time-series datasets with multiple baselines, metrics, and an ablation over the control strength, which makes the evaluation reasonably broad. The method itself is simple and practical: the controller only requires a scalar rescaling of the pretrained score, with no retraining or auxiliary network. The theoretical part provides a fairly complete derivation of stochastic ordering, Wasserstein calibration, and manifold convergence, and the appendix gives detailed proofs and implementation details.
Critical weaknesses
-
Limited novelty and insufficient comparison with related work. The method essentially introduces exponential tilting into an existing likelihood-reweighting framework, which ultimately amounts to a time-dependent score rescaling. The paper does not sufficiently clarify its substantive difference from existing temperature sampling and score-rescaling methods, nor does it provide direct comparisons.
-
A key gap exists between the theoretical target and the actual sampling procedure. Deriving the score of the target distribution at each time step does not guarantee that plugging it into the sampling ODE will generate that target distribution. A standard Gaussian example can illustrate this inconsistency, so the stochastic ordering and Wasserstein calibration results for the target distribution cannot be directly taken as guarantees for the actual algorithm.
-
The experiments do not sufficiently demonstrate the effectiveness of the proposed addition. Training and evaluation use the same set of real data, which limits conclusions about generalization; moreover, the uncontrolled diffusion model already achieves the main ranking results highlighted in the main text. The controller also worsens tail or temporal metrics on some datasets, so the current results do not adequately disentangle the advantage of the base diffusion model from the contribution of the added control strategy.
Questions for authors
-
Can the authors provide a concrete counterexample or condition under which the sampling ODE with per-time score matching does not converge to the target reweighted distribution? Can the paper characterize the conditions (e.g., additional regularity, temporal consistency, or endpoint conditions) under which this construction is valid?
-
What is the essential difference between the controller in Eq. (12) and existing temperature/score-scaling sampling schemes (e.g., multiplying by a constant or time-varying coefficient)? Can the authors provide a direct empirical comparison to isolate the benefit of exponential tilting over simple score rescaling?
-
Since training and evaluation use the same data, can the authors provide results on a held-out test set? In addition, can they report per-dataset comparisons between and , and clearly state on which datasets and metrics the controller provides a significant gain?
3 comments
Thank you for your theoretical response. I have decided to maintain my score.
Clarification of Revisions and Scope
Thank you for the continued discussion. Our revisions are mostly focused on empirical results and numerical experiments, so we believe some miscommunication remains. We would like to provide a clarification of our scope and positioning.
To our knowledge, we are the first to show that a DM can be steered towards low-likelihood samples without specialized training or fine-tuning. The revised experiments compare against the uncontrolled model, DM baselines, and alternative score-scaling schedules, and evaluate on a chronological 80/20 split.
In the Gaussian setting with arbitrary covariance, one can show that holds for all control parameters . For arbitrary distributions, however, this inequality remains open. We do not claim such convergence; rather, our experiments provide empirical support for the inequality (Figures 3 and 4).
As mentioned in our rebuttals, convergence to a -outlier and improving robustness (suggested by reviewer QNZs) are promising directions, but we believe they must be dealt with on a step-by-step basis. The present paper takes a first step by developing and evaluating a practical inference-time mechanism for controllable low-likelihood generation.
We thank the reviewer for the careful theoretical assessment and for recognizing the method’s practicality and breadth of evaluation. Your comments motivated additional diffusion baselines, direct comparisons with other score-scaling schedules, and chronological held-out evaluation. Below, we clarify the scope of our theoretical claims and explain how the expanded experiments isolate the controller’s contribution.
1. Novelty and comparisons with existing methods
At the implementation level, Eq. (12) is a time-dependent score rescaling, motivated by exponential tilting. In our updated experiments (Appendix J), we provided comparisons with the diffusion baselines suggested by Reviewer QNZs: FIDE and HTDM. Particularly, HTDM uses a Student-t-based diffusion formulation for heavy-tail modeling. Our comparison evaluates whether a scalar inference-time modification of a pretrained DM can remain competitive with such specialized diffusion methods. Our method continues to achieve competitive performance, ranking first or second on most evaluation metrics despite requiring no dedicated training.
Appendix K compares constant, linear, exponential, and reciprocal score-scaling schedules in a Gaussian-mixture setting. At the common coefficient λ=0.5, constant and linear schedules are unstable and produce excessive sample magnitudes, while exponential and reciprocal schedules remain stable. The justification of this observation is provided in Appendix K. This comparison demonstrates the importance of the temporal schedule. The stable reciprocal alternative also shows that stable behaviour is achievable with other scaling; however, our exponential scaling is distinguished by being well-motivated by the likelihood-tilting formulation.
2. Target-distribution theory and the actual sampling dynamics
Our theoretical results characterize the intended likelihood-reweighted target and motivate the score modification. They do not guarantee that the sampling dynamics recover this target ρ-outlier measure exactly, and we do not claim such a guarantee. This is left for future work, as stated in the conclusion. We therefore assess whether the implemented sampler exhibits the intended likelihood steering through direct empirical evaluation.
Figure 3 shows that the Wasserstein-1 distance between the generated likelihood distributions and the λ=0 distribution increases monotonically with λ, which is in line with our theory. Figure 4 shows strict leftward shifts in the generated likelihood CDFs as λ increases. Together, these results provide empirical evidence that increasing λ yields samples with progressively lower likelihood under the reference model.
Tables 7–9 in Appendix G provide complementary evidence from the Tail-MSLE metric. Across the tested values of λ, Tail-MSLE decreases as λ increases on every dataset except Weather. Since Tail-MSLE measures agreement with empirical tail magnitudes, this trend suggests that the likelihood shift can also improve tail fidelity, although the effect depends on the dataset.
3. Held-out generalization and the controller’s independent contribution
Appendix J now evaluates all nine datasets using an 80/20 chronological train/test split, with every metric computed against the held-out test period. Tables 11–13 report λ=0 and λ=0.12 side by side, allowing the controller’s contribution to be assessed using the same pretrained model.
As detailed in the table in our response to Reviewer Cu8F, held-out Tail-MSLE95 decreases on seven of nine datasets, with reductions of up to 21.1%. The unweighted average of the dataset-wise percentage reductions is approximately 6.27%, including the deterioration on Weather and the unchanged result on Exchange at reported precision. The tables report variability alongside these descriptive improvements.
We agree that the base DM accounts for much of the strong overall benchmark performance. The controlled-versus-uncontrolled comparisons establish the additional effect of modifying its sampler, while the likelihood CDFs and W1 results demonstrate adjustable outlierness. Dependence and fidelity metrics exhibit trade-offs, so we do not claim that increasing λ improves every metric. The practical result is that a single pretrained model supports increasingly low-likelihood generation and improves Tail-MSLE on several datasets, without another training run.
We hope these clarifications and expanded comparisons address the concerns about theoretical scope, novelty, and empirical attribution, and provide a basis for reconsidering your recommendation.