语言障碍：剖析多语言情境下大语言模型的安全挑战
The Language Barrier: Dissecting Safety Challenges of LLMs in Multilingual Contexts
AI总结：
本文揭示LLMs在低资源语言中更易产生不安全或不相关响应，并指出跨语言对齐瓶颈源于预训练阶段而非指令调优。
摘要：
随着大语言模型（LLMs）的影响力遍及全球各个社区，其在多语言环境中的安全挑战对于对齐研究变得至关重要。本文考察了LLMs在不同语言中所面临安全挑战的差异，并探讨了缓解这些问题的途径。通过比较最先进的LLMs对以高资源语言和低资源语言编写的同一组恶意提示的响应，我们观察到：（1）当恶意提示以低资源语言编写时，LLMs往往更频繁地生成不安全响应；（2）在低资源语言中，LLMs对恶意提示往往生成更多不相关的响应。为了理解这种差异的来源，我们研究了在HH-RLHF数据集上使用基于人类反馈的强化学习（RLHF）或监督微调（SFT）进行指令调优的效果。令人惊讶的是，虽然使用高资源语言进行训练可以改善模型对齐，但在低资源语言中进行训练带来的改进微乎其微。这表明跨语言对齐的瓶颈根植于预训练阶段。我们的发现凸显了跨语言LLM安全方面的挑战，我们希望这些发现能为未来该方向的研究提供参考。
英文摘要：
As the influence of large language models (LLMs) spans across global communities, their safety challenges in multilingual settings become paramount for alignment research. This paper examines the variations in safety challenges faced by LLMs across different languages and discusses approaches to alleviating such concerns. By comparing how state-of-the-art LLMs respond to the same set of malicious prompts written in higher- vs. lower-resource languages, we observe that (1) LLMs tend to generate unsafe responses much more often when a malicious prompt is written in a lower-resource language, and (2) LLMs tend to generate more irrelevant responses to malicious prompts in lower-resource languages. To understand where the discrepancy can be attributed, we study the effect of instruction tuning with reinforcement learning from human feedback (RLHF) or supervised finetuning (SFT) on the HH-RLHF dataset. Surprisingly, while training with high-resource languages improves model alignment, training in lower-resource languages yields minimal improvement. This suggests that the bottleneck of cross-lingual alignment is rooted in the pretraining stage. Our findings highlight the challenges in cross-lingual LLM safety, and we hope they inform future research in this direction.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
