打造17语言安全过滤工具，提升大模型多语种内容安全防护能力
PolyGuard: A Multilingual Safety Moderation Tool for 17 Languages
- 基于191万样本多语言数据训练，覆盖17种语言的通用安全模型
- 在2.9万条多语言测试数据上表现优于现有模型5.5%
- 适合需要多语种内容安全管控的平台与开发者使用
大型语言模型的真实多语言安全监管因局限于少数语言（如英语、中文）及安全定义范围过窄而受阻。为此，我们发布了POLYGUARD——当前最先进的多语言安全模型及其配套训练与评估数据集。POLYGUARD基于包含191万样本的POLYGUARDMIX训练语料库，覆盖17种语言（如中文、捷克语、英语、印地语）。我们还构建了包含2.9万样本的POLYGUARDPROMPTS多语言评测基准，结合真实多语言人机交互与人工验证的机器翻译数据（源自WildGuardMix；Han et al., 2024），包含提示有害性、响应有害性与拒绝响应标签。在多个安全与毒性评测基准上，POLYGUARD性能优于现有开源与商业安全分类器5.5%。本工作推动了面向全球用户的更安全多语言大模型发展。
原文摘要 · Abstract (English)
Truly multilingual safety moderation efforts for Large Language Models (LLMs) have been hindered by a narrow focus on a small set of languages (e.g., English, Chinese) as well as a limited scope of safety definition, resulting in significant gaps in moderation capabilities. To bridge these gaps, we release POLYGUARD, a new state-of-the-art multilingual safety model for safeguarding LLM generations, and the corresponding training and evaluation datasets. POLYGUARD is trained on POLYGUARDMIX, the largest multilingual safety training corpus to date containing 1.91M samples across 17 languages (e.g., Chinese, Czech, English, Hindi). We also introduce POLYGUARDPROMPTS, a high quality multilingual benchmark with 29K samples for the evaluation of safety guardrails. Created by combining naturally occurring multilingual human-LLM interactions and human-verified machine translations of an English-only safety dataset (WildGuardMix; Han et al., 2024), our datasets contain prompt-output pairs with labels of prompt harmfulness, response harmfulness, and response refusal. Through extensive evaluations across multiple safety and toxicity benchmarks, we demonstrate that POLYGUARD outperforms existing state-of-the-art open-weight and commercial safety classifiers by 5.5%. Our contributions advance efforts toward safer multilingual LLMs for all global users.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。