conftrace_

training data privacy

3 papers

Explore in graph

Co-occurring keywords

privacy leakage (105) privacy attack (119) model approximation (11) shadow model (4) risk assessment (78) edit distance (41) membership inference (114) algorithmic recourse (22) membership inference attack (93) property inference attack (3)

Papers

Can Graph Neural Networks Expose Training Data Properties? An Efficient Risk Assessment Approach NIPS 2024
PII-Compass: Guiding LLM training data extraction prompts towards the target PII via grounding ACL 2024
On the Privacy Risks of Algorithmic Recourse AISTATS 2023