Contextualized Privacy Defense for LLM Agents
This addresses privacy risks for users of LLM agents, representing a novel method rather than an incremental improvement.
The paper tackled the problem of limited privacy defenses for LLM agents by proposing Contextualized Defense Instructing (CDI), a new paradigm that uses an instructor model to generate context-aware privacy guidance during execution, resulting in a better balance between privacy preservation (94.2%) and helpfulness (80.6%) compared to baselines.
LLM agents increasingly act on users' personal information, yet existing privacy defenses remain limited in both design and adaptability. Most prior approaches rely on static or passive defenses, such as prompting and guarding. These paradigms are insufficient for supporting contextual, proactive privacy decisions in multi-step agent execution. We propose Contextualized Defense Instructing (CDI), a new privacy defense paradigm in which an instructor model generates step-specific, context-aware privacy guidance during execution, proactively shaping actions rather than merely constraining or vetoing them. Crucially, CDI is paired with an experience-driven optimization framework that trains the instructor via reinforcement learning (RL), where we convert failure trajectories with privacy violations into learning environments. We formalize baseline defenses and CDI as distinct intervention points in a canonical agent loop, and compare their privacy-helpfulness trade-offs within a unified simulation framework. Results show that our CDI consistently achieves a better balance between privacy preservation (94.2%) and helpfulness (80.6%) than baselines, with superior robustness to adversarial conditions and generalization.