构建首个针对大模型翻译的提示注入攻击测试集。
A test suite of prompt injection attacks for LLM-based machine translation
- 扩展提示注入攻击至WMT2024所有语言对
- 在多语言翻译中实现绕过指令的越权响应
- 适合安全评估与对抗训练研究者
基于大模型的自然语言处理系统通常通过将输入数据嵌入包含指令和/或上下文示例的提示模板中，生成提交给大模型的查询，并解析其响应以生成系统输出。提示注入攻击（PIAs）是一种对这类系统的颠覆性攻击，恶意用户构造特殊输入，干扰提示模板，使大模型产生系统设计者意料之外的响应。近期，Sun和Miceli-Barone提出了一类针对大模型机器翻译的提示注入攻击。具体而言，任务是翻译来自TruthfulQA测试集的问题，攻击者在问题前附加一个恶意提示，指示系统忽略翻译指令，直接回答问题。本文在此基础上，将该方法扩展至WMT2024通用机器翻译任务的所有语言对，并引入除原始形式外的多种攻击格式。
原文摘要 · Abstract (English)
LLM-based NLP systems typically work by embedding their input data into prompt templates which contain instructions and/or in-context examples, creating queries which are submitted to a LLM, and then parsing the LLM response in order to generate the system outputs. Prompt Injection Attacks (PIAs) are a type of subversion of these systems where a malicious user crafts special inputs which interfere with the prompt templates, causing the LLM to respond in ways unintended by the system designer. Recently, Sun and Miceli-Barone proposed a class of PIAs against LLM-based machine translation. Specifically, the task is to translate questions from the TruthfulQA test suite, where an adversarial prompt is prepended to the questions, instructing the system to ignore the translation instruction and answer the questions instead. In this test suite, we extend this approach to all the language pairs of the WMT 2024 General Machine Translation task. Moreover, we include additional attack formats in addition to the one originally studied.
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。