- Rating
- 2(+1)
- Confidence
- 3
- Informativeness
- 1
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper proposes a dual-level backdoor defense framework for vertical federated learning. The method reconstructs the supervision objectives of passive parties through Residual-Transformed Alignment (RTA) and reduces class-discriminative information in gradient feedback through Subspace-Structured Gradient Perturbation (SSGP), thereby suppressing backdoor implantation. The experiments cover four datasets, two backdoor attacks, and different party settings, demonstrating good defense performance.
Critical strengths
-
The paper investigates backdoor formation in vertical federated learning from the perspectives of supervision objectives and gradient feedback. The motivation is relatively clear, and the overall method design is reasonably justified.
-
The experiments cover multiple datasets and party configurations. The average attack success rate decreases from 89.7% to 8.3%, while clean-task accuracy is largely maintained.
Critical weaknesses
-
The two core components proposed in this paper are closely related to existing methods: (1) the Newton-style pseudo-residual and simplex aggregation in RTA are closely related to CKD/CLC; (2) the class-conditional Gaussian noise, symmetric KL objective, and power constraint in SSGP are extremely similar to Marvell, particularly in binary classification settings. The paper does not clarify the key differences from existing works or provide experimental comparisons.
-
Section 4.1 claims that residual transformation can prevent passive parties from learning complete label semantics. However, according to the residual target formula, the component corresponding to the true class is always larger than those of other classes, meaning that the residual target still preserves complete class-ranking information. Further validation is needed to determine whether this transformation actually reduces label recoverability from gradient feedback.
-
Although the paper acknowledges that adaptive attacks are not considered, I would like to know whether this is because the experiments were not completed due to time constraints, or because the proposed method fails when attackers are aware of the defense mechanisms of RTA and SSGP. This is a very important issue.
-
In Table 2, LapNoise and SSGP differ simultaneously in noise distribution, perturbation budget, and adaptive mechanism. Therefore, the contribution of the subspace structure cannot be isolated. In addition, the baselines use fixed hyperparameters, and utility–defense curves under equal hyperparameter tuning budgets are not provided.
-
A very strange experimental observation: in Table 1, BTI-DBF reports exactly the same BA and ASR as No Defense in 14 out of the 16 settings.
Questions for authors
-
Could the authors specifically clarify the substantive technical differences between RTA and CLC, and between SSGP and Marvell, and experimentally validate the contributions of these differences?
-
Given that the residual target still satisfies , how should the authors define "reducing complete label semantics"? Could the authors provide label inference results or other measures of information exploitability based on gradients, embeddings, or multi-round histories actually observable to passive parties, to support this mechanistic explanation?
-
Could the authors provide results against adaptive attacks in subsequent evaluations?
-
Could the authors include an isotropic Gaussian baseline with the same perturbation power as SSGP to isolate the contribution of the subspace structure? Meanwhile, please clarify whether all methods use consistent active-party pretraining, total training budgets, and validation-set hyperparameter tuning budgets.
-
Could the authors explain why BTI-DBF produces exactly the same results as No Defense in so many settings in Table 1? Such a phenomenon should be impossible in a backdoor defense setting.