- Rating
- 2
- Confidence
- 2
- Informativeness
- 3
- Presentation
- 2
- Soundness
- 3
Read review
Paper summary
This paper proposes a PGD-like method to modify input images, which makes the victim model produce wrong results when visual token compression is applied.
Critical strengths
- This paper has a comprehensive evaluation under different conditions, and although the method (PGD-like) is naive, it seems to work.
- The loss design is well validated.
Critical weaknesses
- The core contribution seems not enough. This paper seems to be an incremental work compared with the existing work CAA, in that it only adds an explicit semantic preservation loss.
- The background of the problem is unclear; adding a problem formulation section would be better for readers.
- The fragility of this attack method is unclear (no defense methods are used in this paper), such as (i) input purification (Gaussian denoising / median filter / bit-depth reduction), and (ii) a token-level mitigation such as randomized selection or ensembling over compression budgets.
Questions for authors
- Could you please clarify the runtimes of this method and the baselines?
- Could you please provide the performance under different perturbation budgets?