On the Privacy of LLMs: An Ablation Study
Large language models (LLMs) are increasingly deployed in interactive and retrieval-augmented settings, raising significant privacy concerns. While attacks such as Membership Inference (MIA), Attribute Inference (AIA), Data Extraction (DEA), and Backdoor Attacks (BA) have been studied, they are typically analyzed in isolation, leaving a gap in understanding their behavior under common system factors. In this paper, we introduce a unified threat model and notation, reproduce a representative set of privacy attacks, and conduct a structured ablation study to evaluate the impact of key factors such as model architecture, scale, dataset characteristics, and retrieval configuration. Our analysis reveals clear differences across attack types. Membership inference attacks, particularly mask-based variants, exhibit strong and reliable signals, while backdoor attacks achieve consistently high success rates due to their trigger-based nature. In contrast, attribute inference and data extraction attacks remain more challenging, resulting in lower accuracy, yet they pose significant risks as they target sensitive personal information. Overall, these results highlight that privacy risks in LLM systems are highly context-dependent and driven by design choices, emphasizing the need for holistic evaluation and informed deployment practices.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Differentially Private Tabular Data Synthesis using Large Language Models
Synthetic tabular data generation with differential privacy is a crucial problem to enable data sharing with formal privacy. Despite a rich history of methodological research and development, developing differentially pr…
FairnessTabular Data GenerationThe Interlocutor Effect: Why LLMs Leak More Personal Data to Agents Than Humans
Large Language Models (LLMs) alter their privacy behavior based on the perceived identity of their interlocutor. While safety mechanisms typically prevent LLMs from releasing Personally Identifiable Information (PII) to …
1-2-3 Check: Enhancing Contextual Privacy in LLM via Multi-Agent Reasoning
Addressing contextual privacy concerns remains challenging in interactive settings where large language models (LLMs) process information from multiple sources (e.g., summarizing meetings with private and public informat…
Data-Constrained Synthesis of Training Data for De-Identification
Many sensitive domains -- such as the clinical domain -- lack widely available datasets due to privacy risks. The increasing generative capabilities of large language models (LLMs) have made synthetic datasets a viable p…
De-identificationNERCombining Stochastic Defenses to Resist Gradient Inversion: An Ablation Study
Gradient Inversion (GI) attacks are a ubiquitous threat in Federated Learning (FL) as they exploit gradient leakage to reconstruct supposedly private training data. Common defense mechanisms such as Differential Privacy …
Federated LearningImage ClassificationPrivacy Preserving