# GSPR: Aligning LLM Safeguards as Generalizable Safety Policy Reasoners

arXiv: https://arxiv.org/abs/2509.24418 · HTML: https://mlsanity.com/paper/2509.24418/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Sep 2025.

## Built on by (most cited first, 13 of 13)

- [ProGuard: Towards Proactive Multimodal Safeguard](https://mlsanity.com/paper/2512.23573/) (2025)
- [WebAgentGuard: A Reasoning-Driven Guard Model for Detecting Prompt Injection Attacks in Web Agents](https://mlsanity.com/paper/2604.12284/) (2026)
- [Learning Efficient Guardrails for Compliance](https://mlsanity.com/paper/2510.03485/) (2025)
- [Into the Gray Zone: Domain Contexts Can Blur LLM Safety Boundaries](https://mlsanity.com/paper/2604.15717/) (2026)
- [GrandGuard: Taxonomy, Benchmark, and Safeguards for Elderly-Chatbot Interaction Safety](https://mlsanity.com/paper/2605.20203/) (2026)
- [SafePyramid: A Hierarchical Benchmark for In-context Policy Guardrailing](https://mlsanity.com/paper/2606.29887/) (2026)
- [Isolation as a First-Class Principle for LLM-Agent System Safety: Concepts, Taxonomy, Challenges and Future Directions](https://mlsanity.com/paper/2607.12406/) (2026)
- [Towards Multi-modal Multi-turn Safety: From Agentic Interaction to Strategic Alignment](https://mlsanity.com/paper/2601.04736/) (2026)
- [LatentGuard: Efficient and Inspectable Latent Reasoning for LLM Safeguards](https://mlsanity.com/paper/2608.03838/) (2026)
- [Visual Compliance via Executable Safety Rule Entailment](https://mlsanity.com/paper/2609.18328/) (2026)
- [PASTABench: Proactive Assessment of Sequential Trajectories for Agent Safety](https://mlsanity.com/paper/2609.28197/) (2026)
- [Before Agent Tells The Lie: Has Deception Already Been Represented?](https://mlsanity.com/paper/2610.06576/) (2026)
- [Jev-Guard: Fast Routing and Auditing Agent Actions over Editable Policy Trees](https://mlsanity.com/paper/2610.18326/) (2026)

## Abstract (arXiv)

As large language models (LLMs) are integrated into numerous applications, LLMs' safety becomes critical for both application developers and intended users. Currently, great efforts have been made to develop safety benchmarks with fine-grained taxonomies. However, these benchmarks' taxonomies are disparate with different safety policies. Thus, existing safeguards trained on these benchmarks are either coarse-grained to only distinguish between "safe'' and "unsafe,'' or constrained by the specified narrow risk taxonomies. To leverage these fine-grained safety policies across multiple safety taxonomies, we propose GSPR, a Generalizable Safety Policy Reasoner to identify unsafe inputs and outputs with violated safety taxonomies and concise explanations. Unlike prior safeguards which only cover a fixed set of risk factors, GSPR incentivizes its reasoning capability with varied safety taxonomies through reinforcement learning. Our GSPR can be trained across multiple safety benchmarks with distinct taxonomies and naturally exhibits powerful generalization ability. We conduct extensive experiments to show that GSPR significantly improves existing safety guardrails' reasoning capabilities for both safety and category prediction tasks. Moreover, GSPR also achieves the least inference token costs with explanations.