LiSA: Lifelong Safety Adaptation via Conservative Policy Induction
As AI agents move from chat interfaces to systems that read private data, call tools, and execute multi-step workflows, guardrails become a last line of defense against concrete deployment harms. In these settings, guardrail failures are no longer merely answer-quality errors: they can leak secrets, authorize unsafe actions, or block legitimate work. The hardest failures are often contextual: whether an action is acceptable depends on local privacy norms, organizational policies, and user expectations that resist pre-deployment specification. This creates a practical gap: guardrails must adapt to their own operating environments, yet deployment feedback is typically limited to sparse, noisy user-reported failures, and repeated fine-tuning is often impractical. To address this gap, we propose LiSA (Lifelong Safety Adaptation), a conservative policy induction framework that improves a fixed base guardrail through structured memory. LiSA converts occasional failures into reusable policy abstractions so that sparse reports can generalize beyond individual cases, adds conflict-aware local rules to prevent overgeneralization in mixed-label contexts, and applies evidence-aware confidence gating via a posterior lower bound, so that memory reuse scales with accumulated evidence rather than empirical accuracy alone. Across PrivacyLens+, ConFaide+, and AgentHarm, LiSA consistently outperforms strong memory-based baselines under sparse feedback, remains robust under noisy user feedback even at 20% label-flip rates, and pushes the latency--performance frontier beyond backbone model scaling. Ultimately, LiSA offers a practical path to secure AI agents against the unpredictable long tail of real-world edge risks.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Distilled Lifelong Self-Adaptation for Configurable Systems
Modern configurable systems provide tremendous opportunities for engineering future intelligent software systems. A key difficulty thereof is how to effectively self-adapt the configuration of a running system such that …
Soft Actor-Critic-based Control Barrier Adaptation for Robust Autonomous Navigation in Unknown Environments
Motion planning failures during autonomous navigation often occur when safety constraints are either too conservative, leading to deadlocks, or too liberal, resulting in collisions. To improve robustness, a robot must dy…
Autonomous NavigationMotion PlanningSafe Policy Search for Lifelong Reinforcement Learning with Sublinear Regret
Lifelong reinforcement learning provides a promising framework for developing versatile agents that can accumulate knowledge over a lifetime of experience and rapidly learn new tasks by building upon prior knowledge. How…
Lifelong learningreinforcement-learningReinforcement LearningReinforcement Learning (RL)Conservative Offline Policy Adaptation in Multi-Agent Games
Prior research on policy adaptation in multi-agent games has often relied on online interaction with the target agent in training, which can be expensive and impractical in real-world scenarios. Inspired by recent progre…
Dynamic Mixture of Progressive Parameter-Efficient Expert Library for Lifelong Robot Learning
A generalist agent must continuously learn and adapt throughout its lifetime, achieving efficient forward transfer while minimizing catastrophic forgetting. Previous work within the dominant pretrain-then-finetune paradi…
Lifelong learningparameter-efficient fine-tuning