Дифференциалды құпиялылық
Related Papers
Differentially Private Representation for NLP: Formal Guarantee and An Empirical Study on Privacy and Fairness
AbstractIt has been demonstrated that hidden representation learned by deep model can encode private information of the input, hence can be exploited to recover such information with reasonable accuracy. To …...
Empirical Studies of Institutional Federated Learning For Natural Language Processing
AbstractFederated learning has sparkled new interests in the deep learning society to make use of isolated data sources from independent institutes. With the development of novel training tools, we have …...
Certified Robustness to Word Substitution Attack with Differential Privacy
AbstractThe robustness and security of natural language processing (NLP) models are significantly important in real-world applications. In the context of text classification tasks, adversarial examples can be designed by substituting …...
Privacy Regularization: Joint Privacy-Utility Optimization in LanguageModels
AbstractNeural language models are known to have a high capacity for memorization of training samples. This may have serious privacy im- plications when training models on user content such as …...
When differential privacy meets NLP: The devil is in the detail
AbstractDifferential privacy provides a formal approach to privacy of individuals. Applications of differential privacy in various scenarios, such as protecting users’ original utterances, must satisfy certain mathematical properties. Our contribution …...
DP-Rewrite: Towards Reproducibility and Transparency in Differentially Private Text Rewriting
AbstractText rewriting with differential privacy (DP) provides concrete theoretical guarantees for protecting the privacy of individuals in textual documents. In practice, existing systems may lack the means to validate their …...
Just Fine-tune Twice: Selective Differential Privacy for Large Language Models
AbstractProtecting large language models from privacy leakage is becoming increasingly crucial with their wide adoption in real-world products. Yet applying *differential privacy* (DP), a canonical notion with provable privacy guarantees …...
Knowledge Unlearning for Mitigating Privacy Risks in Language Models
AbstractPretrained Language Models (LMs) memorize a vast amount of knowledge during initial pretraining, including information that may violate the privacy of personal lives and identities. Previous work addressing privacy issues …...