- Rating
- 2
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper proposes DAG-FM, a foundation model for causal discovery under heterogeneous causal mechanisms. The method decomposes DAG identification into leaf prediction and parent prediction, implemented by two Transformer modules. A Mixture of Leaf Experts (MoLE) is introduced to accommodate different families of functional causal models (FCMs). The paper also establishes theoretical results on DAG identifiability and evaluates the method on synthetic and real datasets.
Critical strengths
- The decomposition into leaf prediction and parent prediction is intuitive and ensures that the resulting graph is a DAG.
- Considering heterogeneous causal mechanisms is a meaningful direction for causal discovery.
Critical weaknesses
-
The connection between the theoretical results and the proposed method is unclear. My main concern is that the theoretical results are not closely related to the proposed method. The identifiability results characterize the prior space and appear to be applicable to other amortized causal discovery methods under the same assumptions. Moreover, DAG-FM does not explicitly use QPE or enforce the theoretical conditions during training or inference. Although MoLE is motivated by the diversity of FCM families, it is unclear how the theoretical results justify this particular architecture.
-
The theoretical guarantees rely on restrictive assumptions. The identifiability results require additional assumptions, including the finite dimensional representation of QPEs and Assumptions A.2 and A.4. These conditions may limit the applicability of the results to more general causal mechanisms. Moreover, the training prior includes general causal mechanisms for which identifiability has not been established. The discretization procedure also raises questions about whether the smoothness assumptions in the theoretical analysis are satisfied.
-
The experiments do not sufficiently isolate the contribution of the proposed architecture. The synthetic benchmarks are generated using the same prior construction framework as the training data. It is therefore unclear whether the improvements mainly come from the proposed architecture or the choice of training prior. It would be helpful to compare DAG-FM with other amortized causal discovery methods trained on the same prior space.
Questions for authors
- How do the theoretical results justify the specific architecture of DAG-FM?
- To what extent does the actual training prior satisfy the assumptions of Theorem 4.5?
- How would other amortized causal discovery methods perform when trained on the same prior space?
We thank the reviewer for helping us clarify the connection between theory and practice, and for suggesting experimental controls to better isolate the architectural contribution.
See
C1of the global rebuttal. Our contributions are complementary: the theory establishes sufficient conditions for identifiability under heterogeneous mechanisms, while DAG-FM introduces a mechanism-aware architecture for learning across these mechanisms.QPE serves as a proof tool, not an inference component. Its role is to establish when the observational distribution determines a unique DAG within the analyzed prior. A model need not explicitly compute QPE to learn this mapping, just as a statistical identifiability argument need not prescribe an estimation algorithm.
We agree that this result can also inform other amortized methods under the same assumptions. This generality is part of the theoretical contribution, rather than an architectural claim. MoLE provides a separate inductive bias: it explicitly accommodates mechanism heterogeneity through specialization. The theory motivates this setting but does not establish that MoLE is necessary or superior. Its benefit is evaluated experimentally (
C5).The QPE framework allows us to analyze LiNGAM, ANM, HNM, and PNL within a common representation, rather than requiring one homogeneous family throughout the graph. The additional finite-span and non-degeneracy conditions specify where this heterogeneous identifiability result holds.
Before adding General mechanisms and post-processing, the four-family generator matches the theorem’s mechanism-level construction. The remaining conditions concern the allowed PNL functions and the distributional non-degeneracy needed for recursive identification. They are not implied by mechanism-family membership alone.
The broader training prior deliberately extends beyond this analyzed class to improve practical coverage. See
C3for the distinction between the theoretical guarantees and empirical robustness under General mechanisms, discretization, and warping. Thus, the assumptions delimit the guarantee without restricting the empirical study to idealized data.Please see
C5for the complementary control.We again thank the reviewer for the constructive feedback. If these explanations and the additional experimental evidence address your concerns, we would appreciate reconsideration of your assessment.