- Rating
- 3(+1)
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 3
Read review
Paper summary
The authors proposed MultiCFR to estimate treatment effects from tabular, textual, and image covariate. They used a patient-specific tabular anchor to guide the attention over text tokens and image patches. Then they divided the fused representation through a soft gate into a balance-regularized branch and an unregularized predictive branch. They also presented an error-bound decomposition and evaluate their proposed method using two semi-synthetic benchmarks, MIMIC-TIT and IHDP-TIT.
Critical strengths
The main experimental results show improvements over several baselines. The evaluation also includes useful ablation studies on anchors, balancing, and modalities which is very nice to demonstrate the performance of the proposed method.
Critical weaknesses
-
The main novelty of this proposed method lies in combining tabular-conditioned attention with selective representation balancing. However, the trade-off between balancing and prediction is already well established, including in DIGNet, which the paper cites, and the novelty of the two core designs is not sufficiently demonstrated. Direct comparisons are needed with standard tabular conditional attention and disentangled representation methods such as DIGNet, using the same encoder and parameter budget. The current results do not clearly show whether the improvements come from the semantic anchors, general conditional retrieval, or a stronger fusion network.
-
Complementary soft masks and separate projections do not guarantee that the branches learn distinct information. The balanced branch may become nearly constant, while the unbalanced branch captures most predictive information. Moreover, the residual term is not directly controlled during training. This reveals a potential degeneracy and a gap in the theoretical guarantee, but does not establish that collapse occurs in practice. Additionally, some important methods such as DISCRET have been cited but not compared. TransTEE (Zhang, Yi-Fan, et al. "Exploring transformer backbones for heterogeneous treatment effect estimation." arXiv preprint arXiv:2202.01336 (2022).) is another missing baseline model.
Questions for authors
-
Do the two branches learn distinct information? Can you provide diagnostics for branch collapse, treatment prediction, information retention, and residual differences?
-
How does MultiCFR compare with standard covariate-conditioned attention and disentangled representation methods under the same encoders and parameter budgets?
-
Can you report results from the full-tabular oracle and test robustness to different confounding strengths, modality-specific information, and noise levels? Additionally, reporting uncertainty estimates for the main ablations and provide more details on implementation and baseline adaptations would be also helpful?
5 comments
Response [2 / 2]
Q3: Oracle, Robustness, and Uncertainty Evaluation
R3: We added the following additional experiments (Tabs. 7–11): (1) Full-tabular oracle. MultiCFR reduces PEHE by 18.4% compared with the 9-variable tabular baseline, demonstrating its ability to recover useful information from text and images. The Full-25 oracle serves as a full-information reference: its substantially lower error (1.254 vs. 3.121) quantifies the remaining gap between recovering information from transformed multimodal views and directly observing the original covariates.
Tab. 7 Full-tabular Oracle Results
| Factual RMSE | |||
|---|---|---|---|
| Tabular-9 CFR | 3.824±0.616 | 0.505±0.083 | 3.965±0.570 |
| MultiCFR | 3.121±0.545 | 0.430±0.077 | 3.253±0.483 |
| Full-25 CFR | 1.254±0.173 | 0.254±0.037 | 1.681±0.149 |
(2) Confounding strength. MultiCFR consistently outperforms DoubleML-Deep and DIGNet across all confounding strengths (), maintaining the lowest PEHE even under strong confounding and demonstrating robust treatment effect estimation.
Tab.8 Confounding Strength Robustness ()
| Method | ||||
|---|---|---|---|---|
| DoubleML-Deep | 5.710±0.893 | 5.708±0.891 | 5.751±0.894 | 5.783±0.893 |
| DIGNet | 3.074±0.521 | 3.083±0.482 | 3.357±0.554 | 3.659±0.573 |
| MultiCFR | 3.033±0.514 | 3.052±0.505 | 3.240±0.554 | 3.626±0.595 |
(3) Modality-specific confounding. With , we vary the modality driving treatment assignment. MultiCFR consistently achieves the lowest PEHE, outperforming DoubleML-Deep and DIGNet across all settings, demonstrating robustness to modality-specific confounding.
Tab. 9 Robustness to modality-specific information Assignment ()
| Confounding-information source | DoubleML-Deep | DIGNet | MultiCFR |
|---|---|---|---|
| Tabular | 5.747±0.890 | 3.157±0.517 | 3.100±0.513 |
| Text | 5.730±0.892 | 3.186±0.524 | 3.125±0.537 |
| Image | 5.730±0.894 | 3.198±0.549 | 3.105±0.525 |
| All modalities | 5.751±0.894 | 3.357±0.554 | 3.240±0.554 |
(4) Noise robustness. MultiCFR achieves the lowest at all nonzero noise levels. At , its increases by 22.81% relative to the clean setting, compared with 30.25% for DIGNet. MultiCFR also obtains the lowest –noise AUC (3.489), compared with DIGNet (3.518) and DoubleML-Deep (5.710), indicating stronger overall ITE-estimation performance across the noise interval. Although DoubleML-Deep exhibits almost no relative degradation, its consistently high shows that low sensitivity alone does not imply accurate estimation.
Tab. 10 Robustness to Input Noise
| Method | –noise AUC | ||||
|---|---|---|---|---|---|
| DoubleML-Deep | 5.712±0.889 | 5.709±0.888 | 5.708±0.888 | −0.06% | 5.710±0.888 |
| DIGNet | 3.009±0.486 | 3.574±0.568 | 3.919±0.615 | 30.25% | 3.518±0.562 |
| MultiCFR | 3.121±0.545 | 3.524±0.568 | 3.833±0.610 | 22.81% | 3.489±0.568 |
(5) Ablation uncertainty. We report paired 95% confidence intervals for differences. Removing text or images consistently increases , providing statistical evidence for the contribution of multimodal information.
Tab. 11 Ablation Effects with Paired 95% CIs
| Ablation | 95% CI | |
|---|---|---|
| w/o Text | +0.400 | [0.139,0.662] |
| w/o Image | +0.470 | [0.270,0.670] |
| Tabular Only | +0.703 | [0.405,1.001] |
| w/o Anchor | +0.035 | [-0.137,0.206] |
(6) Implementation details. MultiCFR uses frozen BioClinicalBERT and ResNet-18 encoders with cached text and image features, and is trained with AdamW using validation factual MSE for model selection. For fair comparison, all methods use identical multimodal inputs, preprocessing, data splits, and evaluation metrics. When adapting baselines to multimodal ITE estimation, we replace only their input modules while preserving their original mechanisms, including nuisance estimation, representation balancing or disentanglement, propensity modeling, and outcome prediction. We will provide additional implementation and baseline-adaptation details in the revision.
W2: Comparison with DISCRET and TransTEE
R4: We compare MultiCFR with DISCRET and TransTEE, and MultiCFR maintains strong performance.
Tab.12 Comparison with Additional Baselines on MIMIC-TIT
| Method | Factual RMSE | ||
|---|---|---|---|
| TransTEE | 0.443±0.091 | 0.376±0.100 | 1.266±0.005 |
| DISCRET | 0.817±0.058 | 0.738±0.054 | 1.748±0.006 |
| MultiCFR | 0.338±0.032 | 0.120±0.021 | 1.198±0.008 |
Dear Reviewer CPnM,
We greatly appreciate your time and effort in reviewing our work. We hope that our responses have clarified our work and addressed your concerns. If so, we would sincerely appreciate your consideration of updating your rating accordingly. If you have any further concerns, we would be happy to provide further clarification.
We highly value every exchange during the review process and sincerely hope to receive further feedback from you at your earliest convenience.
Sincerely,
The Authors
Response [1 / 2]
We sincerely thank you for your constructive feedback, which has helped us conduct a more comprehensive evaluation. Below are our point-by-point responses.
Q1&W2: Clarification on the Dual-Branch Design
R1: We clarify that MultiCFR employs soft allocation rather than causal disentanglement; thus, the branches need not contain mutually exclusive information. This avoids discarding overlapping confounding and prognostic signals, without suppressing predictive information.
Following your valuable suggestion, we provide collapse, treatment prediction, information retention, and residual diagnostics. The experimental results show that: (1) No branch collapse (Tab. 1): Both branches maintain nonzero feature variance and effective rank, with only 0.67% saturated gates. (2) Treatment prediction (Tab. 2): ZB and ZR achieve nearly identical treatment AUCs (0.691/0.690), confirming that treatment information is not forcibly separated. (3) Information retention (Tab. 3): The joint representation achieves the highest prognostic-score and true-ITE R2 (0.880/0.843), supporting the benefit of soft allocation. (4) Residual discrepancy (Tab. 4): MultiCFR reduces the weighted residual discrepancy from 0.177 to 0.162 (8.5%) and ATE error by 28.1% relative to global balancing, demonstrating effective selective balancing despite information overlap.
Tab. 1 Branch Collapse
Tab. 2 Treatment Prediction
Tab. 3 Information Retention (R2)
Tab. 4 Residual Discrepancy
Q2&W1: Comparison with Standard Attention and Disentangled Representation Methods
R2: Comparison with standard attention. We conducted a controlled 2×2 comparison (Tab. 5), varying only the attention mechanism and balancing strategy while keeping the remaining architecture and training settings consistent. (1) Semantic anchoring: Unlike standard covariate-conditioned attention, our semantic anchor uses explicit tabular semantics to guide patient-specific retrieval of relevant text and image features. Under global balancing, it reduces ϵPEHE by 6.5% and ATE error by 35.8%, validating semantic-guided retrieval. (2) Selective balancing: By balancing only ZB while preserving predictive information, it further reduces ϵPEHE by 2.3%, ATE error by 28.1%, and factual RMSE by 1.6% with semantic anchoring fixed. Together, both designs achieve the best results across all three metrics, demonstrating their complementary benefits beyond a stronger fusion backbone.
Tab. 5 Comparison of Two Key Designs
Values denote ϵPEHE / ϵATE / Factual RMSE.
Comparison with disentangled representations. Under a shared multimodal encoder and comparable parameter budgets, we additionally compared MultiCFR with DR-CFR, DeR-CFR, and DIGNet. As shown in Tab. 6, MultiCFR achieves the lowest ATE error (0.120) and factual RMSE (1.198). Notably, it reduces PEHE by 14.6% and ATE error by 50.8% compared with DIGNet. These results highlight the effectiveness of our anchor-guided soft allocation and selective balancing in preserving predictive information while improving treatment-effect estimation.
Tab. 6 Comparison with Disentangled Representation Methods