Recent studies in large language models (LLMs) shed light on their
multilingual ability and safety, beyond conventional tasks in language
modeling. Still, current benchmarks reveal their inability to comprehensively
evaluate them and are excessively dependent on manual annotations. In this
paper, we introduce code-switching red-teaming (CSRT), a simple yet effective
red-teaming technique that simultaneously tests multilingual understanding and
safety of LLMs. We release the CSRT dataset, which comprises 315 code-switching
queries combining up to 10 languages and eliciting a wide range of undesirable
behaviors. Through extensive experiments with ten state-of-the-art LLMs, we
demonstrate that CSRT significantly outperforms existing multilingual
red-teaming techniques, achieving 46.7% more attacks than existing methods in
English. We analyze the harmful responses toward the CSRT dataset concerning
various aspects under ablation studies with 16K samples, including but not
limited to scaling laws, unsafe behavior categories, and input conditions for
optimal data generation. Additionally, we validate the extensibility of CSRT,
by generating code-switching attack prompts with monolingual data.

最近在大型语言模型（LLM）的研究中，除了传统的语言建模任务，还揭示了它们的多语能力和安全性。本文介绍了一种称为代码混合红队行动（CSRT）的简单而有效的红队行动技术，它同时测试 LLM 的多语理解和安全性。我们发布了 CSRT 数据集，其中包含 315 个代码混合查询，结合多达 10 种语言，并引发了各种不良行为。通过与十种最先进的 LLM 进行广泛实验，我们证明了 CSRT 在英语中比现有方法攻击率高出 46.7%。我们通过对 1.6 万个样本进行消融研究，分析了针对 CSRT 数据集的有害响应，包括但不限于扩展规律、不安全行为类别和最佳数据生成输入条件。此外，通过使用单语数据生成混合代码攻击提示，我们验证了 CSRT 的可扩展性。

CSRT：使用代码切换红队数据集评估和分析 LLMs

CSRT: Evaluation and Analysis of LLMs using Code-Switching Red-Teaming  Dataset

Large language models (LLMs) demonstrate remarkable performance across a
spectrum of languages. In this work, we delve into the question: How do LLMs
handle multilingualism? We introduce a framework that depicts LLMs' processing
of multilingual inputs: In the first several layers, LLMs understand the
question, converting multilingual inputs into English to facilitate the
task-solving phase. In the intermediate layers, LLMs engage in problem-solving
by thinking in English and incorporating multilingual knowledge to obtain
factual content, leveraging the self-attention and feed-forward structures,
respectively. In the last several layers, LLMs generate responses that align
with the original language of the query. In addition, we investigate the
existence of language-specific neurons when processing a certain language. To
detect neurons activated by the input language, even without labels, we
innovatively design a Parallel Language specific Neuron Detection
($\texttt{PLND}$) method that effectively measures the significance of neurons
when handling multilingual inputs. By comprehensive ablation analysis through
deactivating neurons of different layers and structures, we verify the
framework that we propose. Additionally, we demonstrate that we can utilize
such a framework to effectively enhance the multilingual ability with much less
training effort.

大型语言模型在跨多种语言表现出卓越的性能。本文探讨了大型语言模型处理多语言的方式，提出了一个处理多语言输入的框架，并利用该框架验证其有效性，并展示如何通过该框架有效提升多语言能力。