ML-Bench&Guard: Policy-Grounded Multilingual Safety Benchmark and Guardrail for Large Language Models

TL;DR

This paper introduces ML-Bench, a multilingual safety benchmark based on regional regulations, and ML-Guard, a guardrail model that supports policy-conditioned safety assessment across 14 languages.

Contribution

It presents a new policy-grounded multilingual safety benchmark and a diffusion-based guardrail model that outperforms existing methods in safety evaluation.

Findings

ML-Guard outperforms 11 strong baselines across 6 benchmarks.

ML-Bench enables culturally and legally aligned safety evaluation.

Two ML-Guard variants support fast safety checks and detailed compliance assessment.

Abstract

As Large Language Models (LLMs) are increasingly deployed in cross-linguistic contexts, ensuring safety in diverse regulatory and cultural environments has become a critical challenge. However, existing multilingual benchmarks largely rely on general risk taxonomies and machine translation, which confines guardrail models to these predefined categories and hinders their ability to align with region-specific regulations and cultural nuances. To bridge these gaps, we introduce ML-Bench, a policy-grounded multilingual safety benchmark covering 14 languages. ML-Bench is constructed directly from regional regulations, where risk categories and fine-grained rules derived from jurisdiction-specific legal texts are directly used to guide the generation of multilingual safety data, enabling culturally and legally aligned evaluation across languages. Building on ML-Bench, we develop ML-Guard, a Diffusion Large Language Model (dLLM)-based guardrail model that supports multilingual safety judgment and policy-conditioned compliance assessment. ML-Guard has two variants, one 1.5B lightweight model for fast `safe/unsafe' checking and a more capable 7B model for customized compliance checking with detailed explanations. We conduct extensive experiments against 11 strong guardrail baselines across 6 existing multilingual safety benchmarks and our ML-Bench, and show that ML-Guard consistently outperforms prior methods. We hope that ML-Bench and ML-Guard can help advance the development of regulation-aware and culturally aligned multilingual guardrail systems.