<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="miceli-barone-sun-2024-test">
    <titleInfo>
        <title>A Test Suite of Prompt Injection Attacks for LLM-based Machine Translation</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Antonio</namePart>
        <namePart type="given">Valerio</namePart>
        <namePart type="family">Miceli Barone</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Zhifan</namePart>
        <namePart type="family">Sun</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2024-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the Ninth Conference on Machine Translation</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Barry</namePart>
            <namePart type="family">Haddow</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Tom</namePart>
            <namePart type="family">Kocmi</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Philipp</namePart>
            <namePart type="family">Koehn</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Christof</namePart>
            <namePart type="family">Monz</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Miami, Florida, USA</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>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, then parse 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 interfer with the prompt templates, causing the LLM to respond in ways unintended by the system designer.Recently, Sun and Miceli-Barone (2024) 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.</abstract>
    <identifier type="citekey">miceli-barone-sun-2024-test</identifier>
    <location>
        <url>https://aclanthology.org/2024.wmt-1.30</url>
    </location>
    <part>
        <date>2024-11</date>
        <extent unit="page">
            <start>380</start>
            <end>450</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T A Test Suite of Prompt Injection Attacks for LLM-based Machine Translation
%A Miceli Barone, Antonio Valerio
%A Sun, Zhifan
%Y Haddow, Barry
%Y Kocmi, Tom
%Y Koehn, Philipp
%Y Monz, Christof
%S Proceedings of the Ninth Conference on Machine Translation
%D 2024
%8 November
%I Association for Computational Linguistics
%C Miami, Florida, USA
%F miceli-barone-sun-2024-test
%X 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, then parse 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 interfer with the prompt templates, causing the LLM to respond in ways unintended by the system designer.Recently, Sun and Miceli-Barone (2024) 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.
%U https://aclanthology.org/2024.wmt-1.30
%P 380-450

Markdown (Informal)

[A Test Suite of Prompt Injection Attacks for LLM-based Machine Translation](https://aclanthology.org/2024.wmt-1.30) (Miceli Barone & Sun, WMT 2024)

• A Test Suite of Prompt Injection Attacks for LLM-based Machine Translation (Miceli Barone & Sun, WMT 2024)

ACL

• Antonio Valerio Miceli Barone and Zhifan Sun. 2024. A Test Suite of Prompt Injection Attacks for LLM-based Machine Translation. In Proceedings of the Ninth Conference on Machine Translation, pages 380–450, Miami, Florida, USA. Association for Computational Linguistics.