- Rating
- 2
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper investigates what models learn from unlearnable examples (UEs), beyond their low accuracy on clean test data. Through shallow-network probes, perturbation-transfer experiments, and output-label remapping, the authors distinguish two observed behaviors: perturbation coding and image–perturbation coupling. Motivated by these findings, they formulate a decodable perturbation shortcut (DPS) objective and propose LCSC, which jointly optimizes class-shared perturbations and a periodically reinitialized reader, together with a factorized variant, LCSC-LR. Experiments demonstrate protection across multiple datasets and architectures, with LCSC-LR providing improved resistance to JPEG compression and small-radius adversarial training.
Critical strengths
-
The diagnostics distinguish perturbation-driven predictions from image-dependent behavior. Label remapping shows that low clean accuracy can conceal recoverable class information.
-
LCSC-LR outperforms the tested baselines in most JPEG settings and under small-radius adversarial training. Construction is also faster than TAP, SEP, and REM.
Critical weaknesses
-
Output-level diagnostics do not establish whether semantic information remains in intermediate features. Table 7 also shows that removing cross-class pairing or reader resets does not weaken protection at 8/255, leaving their contributions unclear.
-
Final accuracy can overstate protection: early stopping recovers 65.3% accuracy for LCSC at 1/255. Most countermeasure comparisons use only CIFAR-10/ResNet-18 and a single seed.
Questions for authors
-
Can linear probes on frozen features determine whether semantic information remains at early and final checkpoints?
-
Under matched construction budgets, when do cross-class pairing and reader resets improve over class-aligned optimization and class-wise EM?
-
Do the JPEG and adversarial-training gains persist across multiple seeds and another dataset or architecture?
-
Why does low-rank parameterization improve JPEG resistance? Can decoding accuracy before and after compression explain the difference between LCSC and LCSC-LR?
-
Does LCSC remain effective when only part of the training set is protected and the learner also has access to unprotected samples?