RePolicy: Reinforcement Learning for Safety-Policy Invocation in Agent Safeguards

• URL: http://arxiv.org/abs/2608.24275v2
• Date: Thu, 27 Aug 2026 06:48:15 GMT
• Title: RePolicy: Reinforcement Learning for Safety-Policy Invocation in Agent Safeguards
• Abstract summary: We propose RePolicy, an agent safeguard that learns safety-policy invocation through reinforcement learning.<n>Given an agent trajectory and a dynamic policy library, RePolicy invokes the applicable policy and uses its content to produce a policy-grounded rationale and safety judgment.
• Score: 31.437761312893667
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Safeguarding language model agents requires assessing complete execution trajectories under context-dependent safety policies. Existing policy-aware safeguards mainly rely on prompting or supervised fine-tuning, limiting their ability to adapt to unseen trajectories and changing policy contexts. We propose RePolicy, an agent safeguard that learns safety-policy invocation through reinforcement learning. Given an agent trajectory and a dynamic policy library, RePolicy invokes the applicable policy and uses its content to produce a policy-grounded rationale and safety judgment. We construct PolicyTraj-20K to support supervised initialization, followed by GRPO with verifiable rewards and policy-context perturbation. Experiments across six agent safety benchmarks show that RePolicy achieves strong overall safety-detection performance and robust policy invocation under varying policy contexts.

Related papers

• PolicyMem: Geometric Policy Memory for LLM Governance [72.04962480384322]
  We introduce PolicyMem, a geometric policy memory that externalizes natural-language policies as reusable geometric memory objects.<n>A memory writer compiles natural-language policies into policy memory slots, and query-response pairs read the policy memory through projection energy.<n>The resulting policy-evidence profile directly mediates the safety verdict and is reused for policy attribution and post-intervention verification.
  arXiv  Detail & Related papers  (2026-09-12T06:07:43Z)
• PolicyAlign: Direct Policy-Based Safety Alignment for Large Language Models [19.807112285121715]
  In real-world deployment, emerging safety requirements are often specified as natural-language policies.<n>This creates a mismatch between rapidly evolving safety policies and conventional data-driven alignment methods.<n>We propose PolicyAlign, a framework for directly aligning large language models with safety policies.
  arXiv  Detail & Related papers  (2026-06-24T06:10:33Z)
• Beyond Static Alignment: Hierarchical Policy Control for LLM Safety via Risk-Aware Chain-of-Thought [5.251527748612469]
  Large Language Models (LLMs) face a fundamental safety-helpfulness trade-off due to static, one-size-fits-all safety policies.<n>We present textbfPACT (Prompt-Thought Action via Chain-of-Thought), a framework for dynamic safety control through explicit, risk-aware reasoning.
  arXiv  Detail & Related papers  (2026-02-06T12:20:01Z)
• A universal policy wrapper with guarantees [0.0]
  We introduce a universal policy wrapper for reinforcement learning agents.<n>Our wrapper selectively switches between a high-performing base policy and a fallback policy.<n>It operates without needing additional system knowledge or online constrained optimization.
  arXiv  Detail & Related papers  (2025-05-18T10:37:27Z)
• SPoRt -- Safe Policy Ratio: Certified Training and Deployment of Task Policies in Model-Free RL [54.022106606140774]
  We present theoretical results that place a bound on the probability of violating a safety property for a new task-specific policy in a model-free, episodic setting.<n>This bound can be applied to temporally-extended properties (beyond safety) and to robust control problems.<n>We present experimental results demonstrating this trade-off and comparing the theoretical bound to posterior bounds derived from empirical violation rates.
  arXiv  Detail & Related papers  (2025-04-08T19:09:07Z)
• Safety Correction from Baseline: Towards the Risk-aware Policy in Robotics via Dual-agent Reinforcement Learning [64.11013095004786]
  We propose a dual-agent safe reinforcement learning strategy consisting of a baseline and a safe agent. Such a decoupled framework enables high flexibility, data efficiency and risk-awareness for RL-based control. The proposed method outperforms the state-of-the-art safe RL algorithms on difficult robot locomotion and manipulation tasks.
  arXiv  Detail & Related papers  (2022-12-14T03:11:25Z)
• Bounded Robustness in Reinforcement Learning via Lexicographic Objectives [54.00072722686121]
  Policy robustness in Reinforcement Learning may not be desirable at any cost. We study how policies can be maximally robust to arbitrary observational noise. We propose a robustness-inducing scheme, applicable to any policy algorithm, that trades off expected policy utility for robustness.
  arXiv  Detail & Related papers  (2022-09-30T08:53:18Z)
• Boolean Decision Rules for Reinforcement Learning Policy Summarisation [16.969788244589388]
  We create a rule-based summary of an agent's policy using a lava gridworld. We discuss possible avenues to introduce safety into a RL agent's policy by using rules generated by this rule-based model as constraints imposed on the agent's policy.
  arXiv  Detail & Related papers  (2022-07-18T14:51:24Z)
• Safe Reinforcement Learning via Confidence-Based Filters [78.39359694273575]
  We develop a control-theoretic approach for certifying state safety constraints for nominal policies learned via standard reinforcement learning techniques. We provide formal safety guarantees, and empirically demonstrate the effectiveness of our approach.
  arXiv  Detail & Related papers  (2022-07-04T11:43:23Z)
• SAFER: Data-Efficient and Safe Reinforcement Learning via Skill Acquisition [59.94644674087599]
  We propose SAFEty skill pRiors (SAFER), an algorithm that accelerates policy learning on complex control tasks under safety constraints. Through principled training on an offline dataset, SAFER learns to extract safe primitive skills. In the inference stage, policies trained with SAFER learn to compose safe skills into successful policies.
  arXiv  Detail & Related papers  (2022-02-10T05:43:41Z)
• Learning Goal-oriented Dialogue Policy with Opposite Agent Awareness [116.804536884437]
  We propose an opposite behavior aware framework for policy learning in goal-oriented dialogues. We estimate the opposite agent's policy from its behavior and use this estimation to improve the target agent by regarding it as part of the target policy.
  arXiv  Detail & Related papers  (2020-04-21T03:13:44Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.