Can Sensitive Information Be Deleted From LLMs? Objectives for Defending Against Extraction Attacks.
Vaidehi PatilPeter HaseMohit BansalPublished in: CoRR (2023)
Keyphrases
- sensitive information
- information leakage
- malicious users
- privacy preserving
- third party
- ddos attacks
- privacy preservation
- data privacy
- privacy protection
- sensitive data
- original data
- private information
- privacy issues
- data publishing
- medical records
- confidential information
- data protection
- privacy preserving data publishing
- information extraction
- watermarking algorithm
- personal information
- personal data
- private data
- classification rules
- feature space
- pattern recognition
- support vector
- neural network