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

arXiv:2608.24275 · cs.AI, cs.CL · Submitted 2026-08-25 · Read on arXiv

cs.AI, cs.CL

Submitted: 2026-08-25

Updated: 2026-08-27

Code: https://github.com/jianghoucheng/RePolicy

License: http://creativecommons.org/licenses/by/4.0/

The gist: Safeguarding language model agents requires assessing complete execution trajectories under context-dependent safety policies.

Terminology

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.

Sources

Related papers