arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.14016cs.CL
收藏

LLM 作为评判者是否稳健?探究零样本 LLM 评估上的通用对抗攻击

Is LLM-as-a-Judge Robust? Investigating Universal Adversarial Attacks on Zero-shot LLM Assessment

University of Cambridge(剑桥大学)

Vyas Raina, Adian Liusie, Mark Gales

AI总结:

本文首次研究 LLM 评估器的对抗鲁棒性,提出通过代理模型生成可迁移的简短通用对抗短语,使未知评判模型在绝对评分中给出虚高分数,揭示了 LLM-as-a-judge 方法的可靠性风险。

摘要:

大型语言模型(LLMs)是强大的零样本评估器,被用于现实场景中,例如评估书面考试和基准测试系统。尽管这些应用至关重要,但尚无现有工作分析评判型 LLM 对对抗操纵的脆弱性。本工作首次研究了评估型 LLM 的对抗鲁棒性,我们证明可以将简短的通用对抗短语拼接起来,欺骗评判型 LLM 预测虚高的分数。由于对手可能不知道或无法访问评判型 LLM,我们提出一种简单的代理攻击方法:首先攻击一个代理模型,然后将学习到的攻击短语迁移到未知的评判型 LLM 上。我们提出一种实用算法来确定简短的通用攻击短语,并证明当迁移到未见过的模型时,分数可以被大幅抬高,以至于无论被评估文本如何,都会预测出最高分。研究发现,与比较式评估相比,评判型 LLM 在用于绝对评分时对这些对抗攻击的敏感性显著更高。我们的发现引发了对 LLM 作为评判者方法可靠性的担忧,并强调在高风险现实场景部署之前,解决 LLM 评估方法中脆弱性的重要性。

英文摘要:

Large Language Models (LLMs) are powerful zero-shot assessors used in real-world situations such as assessing written exams and benchmarking systems. Despite these critical applications, no existing work has analyzed the vulnerability of judge-LLMs to adversarial manipulation. This work presents the first study on the adversarial robustness of assessment LLMs, where we demonstrate that short universal adversarial phrases can be concatenated to deceive judge LLMs to predict inflated scores. Since adversaries may not know or have access to the judge-LLMs, we propose a simple surrogate attack where a surrogate model is first attacked, and the learned attack phrase then transferred to unknown judge-LLMs. We propose a practical algorithm to determine the short universal attack phrases and demonstrate that when transferred to unseen models, scores can be drastically inflated such that irrespective of the assessed text, maximum scores are predicted. It is found that judge-LLMs are significantly more susceptible to these adversarial attacks when used for absolute scoring, as opposed to comparative assessment. Our findings raise concerns on the reliability of LLM-as-a-judge methods, and emphasize the importance of addressing vulnerabilities in LLM assessment methods before deployment in high-stakes real-world scenarios.

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。

↑