一篇新研究论文介绍了一种针对东亚和东南亚文化背景的大型语言模型（LLMs）进行文化适应性红队测试的方法学。研究发现，直接翻译英文基准测试会严重低估LLM的风险，而经过文化适应性调整的提示语能带来更高的攻击成功率。该研究强调，安全评估有必要根据特定的文化细微差别进行调整，而不是仅仅依赖语言翻译。 AI
影响 将LLM安全评估调整至文化背景至关重要，以实现可靠的多语言部署。
排序理由 该集群包含一篇学术论文，详细介绍了评估LLM安全性的新方法。
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
arXiv:2606.09178v1 Announce Type: cross Abstract: Multilingual safety evaluation of large language models (LLMs) has predominantly relied on direct translation (DT) of English benchmarks into target languages - an approach that converts surface-level linguistic form while failing…
Multilingual safety evaluation of large language models (LLMs) has predominantly relied on direct translation (DT) of English benchmarks into target languages - an approach that converts surface-level linguistic form while failing to reflect the cultural context embedded in threa…
Multilingual safety evaluation of large language models (LLMs) has predominantly relied on direct translation (DT) of English benchmarks into target languages - an approach that converts surface-level linguistic form while failing to reflect the cultural context embedded in threa…