- Rating
- 3(+1)
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
The paper introduces a new approach to Causal Discovery ensembling that takes into account various Causal Discovery method outputs as well as observational data statistics. It evaluates five ensembling variants with increasing complexity and compares them against individual Causal Discovery methods, an alternative ensembling approach and ammortized Causal Discovery.
Critical strengths
Ensembling of Causal Discovery methods is an interesting approach, especially with the help of observational statistics. The presented method is timely and provides a meaningful alternative to other ensembling or ammortized approaches to Causal Discovery.
The approach of testing the 5 different levels of ensembling is rigorous and comparing them all with the baseline Causal Discovery methods is solid. I also like the clear six data partitions with clear roles.
Generally the manuscript is very detailed and provides a lot of information on the specific design of the proposed method.
Critical weaknesses
While the paper is on a good way, however I have a number of concerns that I list below (not ordered by importance):
-
There are many design choices in this method that are not really explained. Examples for this are the network architecture (Why 1/2 layers? Why exactly these 8 statistical data features? δ_max and g_max. Was there HP optimization involved? A general discussion on the design philosophy of the approach would be great. Exemplary, the candidate-budget ablation (G.15) is solid.
-
The literature review is rather sparse. Specifically I am missing 2 additional ensembling approaches for Causal Discovery that I am aware of [1][2] and some more recent amortized approaches, e.g. [3][4].
-
The paper is very dense and has no figures whatsoever. A depiction of the proposed ensembling approach and its components would likely help with comprehensibility.
-
"Full regularizer definitions and coefficients are provided in Appendix E": Appendix E exists but is missing content.
-
The real world data evaluation is quite sparse. An additional dataset would make it much easier to asses the performance of the approach out-of-distribution.
-
Finally a more fundamental concern that is partly addressed already in the text: As all learned components are only trained on a very narrow training distribution, what is the reasoning why these learned modules should behave in a meaningful manner on out-of-distribution data? The results on Sachs seem to support this issue. For new data, the conditional patterns that your method can rely on will change. E.g. if no module will ever see non-gaussian noise during training, it will not like not trust noise-based approaches like LINGAM. While you noted this as future work, some more discussion on this topic would be great!
[1] Stein, Gideon, et al. "TCD-Arena: Assessing Robustness of Time Series Causal Discovery Methods Against Assumption Violations." arXiv preprint arXiv:2605.03045 (2026).
[2] Wu, Menghua, et al. "Sample, estimate, aggregate: A recipe for causal discovery foundation models." arXiv preprint arXiv:2402.01929 (2024).
[3] Blöbaum, Patrick, Krishnakumar Balasubramanian, and Shiva Prasad Kasiviswanathan. "FoundCause: Causal Discovery with Latent Confounders from Observational Data." arXiv preprint arXiv:2606.17516 (2026).
[4] Guan, Zhengkang, et al. "DCD-PFN: A Decoupling-Aware Foundation Model for Causal Discovery." arXiv preprint arXiv:2606.21212 (2026).
Questions for authors
Could you comment on your choice of relying on an edge independent weighting instead of a joint estimation of all links as in [1] ? What could be (dis)advantages?
What prevented you from training on a much broader synthetic distribution to alleviate the sym-to-real gap at least partially?
11 comments
W1 – What motivated the architecture, feature set, and bounded design choices?
CaFuFo was designed as a sequence of controlled, low-capacity extensions to candidate voting rather than as a large unconstrained predictor. Each stage relaxes one limitation of the preceding stage: global weights capture average constituent reliability; a single self-attention layer enables edge-specific interactions among candidates; a bounded logit correction introduces limited observation-derived evidence; an independent Data-Only Expert permits predictions beyond candidate support; and the router controls how much of this less-constrained expert contributes to the final prediction. The design therefore increases flexibility progressively while deliberately preserving candidate evidence.
The architecture and bounds were fixed during pre-lock development and were not selected using labels from the locked evaluation tasks, topology-transfer tasks, or Sachs. We did not conduct a broad automated architecture or hyperparameter sweep. Model parameters were fitted on training tasks, the stopping epoch was selected using validation tasks, and temperature and graph threshold were selected using the separate calibration split.
Table 11 summarizes the main architectural choices and their provenance. The bounded components have direct operational interpretations. The reliability adjustment changes a candidate's prior contribution by at most a multiplicative factor between and , rather than permitting unrestricted replacement of the prior. The Candidate Anchor correction changes edge odds by at most a factor of . Finally, ensures that the full hybrid cannot completely discard candidate-derived evidence. We regard these values as conservative safeguards rather than empirically optimized constants.
| Choice | Value | Design rationale | Selection |
|---|---|---|---|
| Hidden width / heads | Compact representation with multiple candidate-interaction channels | Fixed pre-lock | |
| Candidate-transformer depth | 1 layer | One interaction step models candidate agreement and conflict while limiting capacity | Fixed pre-lock |
| Data-expert depth | 2 layers | Slightly greater capacity because this branch lacks candidate-graph scaffolding | Fixed pre-lock |
| Pair statistics | 8 | Compact observation-derived description of pairwise association, directional asymmetry, marginal structure, and task size | Hand specified |
| Reliability bound | Prevents edge-specific reliability adjustments from departing arbitrarily from the global candidate prior | Fixed conservative bound | |
| Correction bound | Limits the Candidate Anchor correction to one logit unit | Fixed conservative bound | |
| Maximum data gate | Guarantees at least Candidate Anchor contribution | Fixed conservative bound | |
| Stopping epoch | At most | Controls optimization without changing model capacity | Validation selected |
| Temperature / threshold | One per method and seed | Calibrates probabilities and extracts the final DAG | Calibration selected |
Table 11. Rationale and provenance of the principal architectural and optimization choices.
W2 – Coverage of related work.
We thank the reviewer for identifying these omissions. We have expanded the related-work section to discuss SEA and its sample-estimate-aggregate framework, the ensemble robustness analysis in TCD-Arena, and the recent amortized discovery models FoundCause and DCD-PFN.
Appreciation of Review (93f7)
We thank the reviewer for recognizing the timeliness of combining heterogeneous causal discovery methods with observational statistics, the value of our progressive five-variant evaluation, the breadth of the baseline comparisons, and the clearly separated roles of the six data partitions. We also appreciate their positive assessment of the manuscript’s technical detail.
Changes in the revised manuscript (as a result of your review) can be seen in cyan.