Vulnerability vs. Reliability: Disentangled Adversarial Examples for Cross-Modal Learning
Explore this paper's citation graph
Summary
Novel Disentangled Adversarial examples for Cross-Modal learning is proposed, dubbed DACM, which applies the generation of adversarial perturbations to strengthen cross-modal correlations, wherein the modality-related component is acquired through gradually detaching the modalities-unrelated component.
- Type
- article
- Published
- 2020-07-06
- Cited by
- 22
- References
- 58
- OpenAlex
- https://openalex.org/W3080270960
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:221191712
Keywords
Modal, Adversarial system, Computer science, Modality (human–computer interaction), Deep learning
References
- Intriguing properties of neural networks
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Supervised Discrete Hashing
- Semantics-preserving hashing for cross-view retrieval
- Deep neural networks are easily fooled: High confidence predictions for unrecognizable images
- Supervised hashing with kernels
- NUS-WIDE: a real-world web image database from National University of Singapore
- Im2Text: Describing Images Using 1 Million Captioned Photographs
- On the Role of Correlation and Abstraction in Cross-Modal Multimedia Retrieval
- Discrete Graph Hashing
- Compact Hyperplane Hashing with Bilinear Functions
- The MIR flickr retrieval evaluation
- Active Object Localization with Deep Reinforcement Learning
- Deep Residual Learning for Image Recognition
- DeepFool: A Simple and Accurate Method to Fool Deep Neural Networks
- Deep Cross-Modal Hashing
- Deep Visual-Semantic Hashing for Cross-Modal Retrieval
- TensorFlow: a system for large-scale machine learning
- Generative Adversarial Text to Image Synthesis
- Adversarial examples in the physical world
Cited by
- Multimodal graph inference network for scene graph generation
- Prototype-supervised Adversarial Network for Targeted Attack of Deep Hashing
- Deep Momentum Uncertainty Hashing
- Adversarial Examples Generation for Deep Product Quantization Networks on Image Retrieval
- Proactive Privacy-preserving Learning for Cross-modal Retrieval
- A Hierarchical Graph Learning Model for Brain Network Regression Analysis
- Efficient Query-based Black-box Attack against Cross-modal Hashing Retrieval
- Rethinking Label Flipping Attack: From Sample Masking to Sample Thresholding
- Robust Cross-Modal Retrieval by Adversarial Training
- Targeted Adversarial Attack Against Deep Cross-Modal Hashing Retrieval
- Advancing Adversarial Training by Injecting Booster Signal
- Invisible Black-Box Backdoor Attack against Deep Cross-Modal Hashing Retrieval
- Primary Code Guided Targeted Attack against Cross-modal Hashing Retrieval
- Diversified perturbation guided by optimal target code for cross-modal adversarial attack
- Modality-Specific Interactive Attack for Vision-Language Pre-Training Models
- FPAD: Fuzzy-Prototype-Guided Adversarial Attack and Defense for Deep Cross-Modal Hashing
- Gradient Pruning Interactive Attack for Vision-Language Pre-Training Models
- Cross-Gen: An Efficient Generator Network for Adversarial Attacks on Cross-Modal Hashing Retrieval
- ReGeNet: Relevance-Guided Generative Network to Evaluate the Adversarial Robustness of Cross-Modal Retrieval Systems
- Understanding and Measuring Robustness of Multimodal Learning