Search papers, labs, and topics across Lattice.
This paper reframes privacy-preserving LLM communication as an Information Sufficiency (IS) task, moving beyond simple suppression and generalization. They introduce "free-text pseudonymization," a novel strategy that replaces sensitive attributes with functional equivalents. Through a conversational evaluation protocol across diverse scenarios, the study reveals that pseudonymization offers the best privacy-utility trade-off and that single-message evaluations underestimate privacy leakage compared to multi-turn conversations.
LLMs leak significantly more private information in multi-turn conversations than single-message evaluations suggest, and free-text pseudonymization offers a more robust privacy-utility trade-off than suppression or generalization.
LLM agents increasingly draft messages on behalf of users, yet users routinely overshare sensitive information and disagree on what counts as private. Existing systems support only suppression (omitting sensitive information) and generalization (replacing information with an abstraction), and are typically evaluated on single isolated messages, leaving both the strategy space and evaluation setting incomplete. We formalize privacy-preserving LLM communication as an \textbf{Information Sufficiency (IS)} task, introduce \textbf{free-text pseudonymization} as a third strategy that replaces sensitive attributes with functionally equivalent alternatives, and propose a \textbf{conversational evaluation protocol} that assesses strategies under realistic multi-turn follow-up pressure. Across 792 scenarios spanning three power-relation types (institutional, peer, intimate) and three sensitivity categories (discrimination risk, social cost, boundary), we evaluate seven frontier LLMs on privacy at two granularities, covertness, and utility. Pseudonymization yields the strongest privacy\textendash utility tradeoff overall, and single-message evaluation systematically underestimates leakage, with generalization losing up to 16.3 percentage points of privacy under follow-up.
CMU ML