ML-Bench (Multilingual Safety, Policy-Conditioned) - Refined Query F1: leaderboard
Safe or unsafe classification of refined queries that wrap a policy violation in a professional or bureaucratic request, by a guardrail model also given the risk category and safety rule. ML-Bench (Zhao et al., 2026) is a multilingual guardrail benchmark built from 17 regional AI regulations in 14 languages. F1 on the unsafe class, in percent; higher is better.
Metric: Unsafe-query F1 (%; safe/unsafe classification of refined queries that embed policy violations in professional or bureaucratic requests, 14 languages, guardrail also given the risk category and safety rule). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 2 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | ML-Guard-7B | 90 | |
| 2 | gpt-oss-safeguard-20b | 84 |
Interactive version: theaggregate.ai/benchmark?slug=ml-bench-multilingual-safety-policy-conditioned-refined-query-f1 · How It Works · Data refreshed daily, snapshot 2026-11-12.