利用目标对齐的对抗性攻击有效伪造言语欺骗检测
Effective faking of verbal deception detection with target-aligned adversarial attacks
AI总结：
本研究提出利用大型语言模型进行目标对齐的对抗性攻击以重写欺骗性陈述，发现当攻击与评估目标对齐时，人类和机器的欺骗检测能力均降至随机水平，揭示了现有检测方法的脆弱性。
摘要：
背景：通过分析语言进行欺骗检测是一条极具前景的途径，该过程同时利用了人类判断和自动化机器学习判断。对于这两种形式的可信度评估，将欺骗性陈述重写为看似真实的自动化对抗性攻击构成了严重威胁。方法：我们在一项针对人类和机器学习模型的欺骗检测任务中，使用了一个包含 243 个真实和 262 个捏造的自传体故事的数据集。我们利用大型语言模型来重写欺骗性陈述，使其看起来像真实的。在研究 1 中，做出欺骗判断或使用详细性启发式的人类，以及两个机器学习模型（一个微调语言模型和一个简单的 n-gram 模型），对欺骗性陈述的原始版本或对抗性修改版本进行判断。在研究 2 中，我们操纵了修改版本的目标对齐，即根据陈述是由人类还是计算机模型评估来定制攻击。结果：当对抗性修改与其目标对齐时，人类（d=-0.07 和 d=-0.04）和机器判断（51% 准确率）均降至随机水平。当攻击未与目标对齐时，人类启发式判断（d=0.30 和 d=0.36）和机器学习预测（63-78%）均显著优于随机水平。结论：易于获取的语言模型能够有效帮助任何人伪造欺骗检测工作，从而规避人类和机器学习模型的检测。人类和机器对对抗性修改的鲁棒性取决于该目标对齐。我们最后提出了利用对抗性攻击设计和技术推进欺骗研究的建议。
英文摘要：
Background: Deception detection through analysing language is a promising avenue using both human judgments and automated machine learning judgments. For both forms of credibility assessment, automated adversarial attacks that rewrite deceptive statements to appear truthful pose a serious threat. Methods: We used a dataset of 243 truthful and 262 fabricated autobiographical stories in a deception detection task for humans and machine learning models. A large language model was tasked to rewrite deceptive statements so that they appear truthful. In Study 1, humans who made a deception judgment or used the detailedness heuristic and two machine learning models (a fine-tuned language model and a simple n-gram model) judged original or adversarial modifications of deceptive statements. In Study 2, we manipulated the target alignment of the modifications, i.e. tailoring the attack to whether the statements would be assessed by humans or computer models. Results: When adversarial modifications were aligned with their target, human (d=-0.07 and d=-0.04) and machine judgments (51% accuracy) dropped to the chance level. When the attack was not aligned with the target, both human heuristics judgments (d=0.30 and d=0.36) and machine learning predictions (63-78%) were significantly better than chance. Conclusions: Easily accessible language models can effectively help anyone fake deception detection efforts both by humans and machine learning models. Robustness against adversarial modifications for humans and machines depends on that target alignment. We close with suggestions on advancing deception research with adversarial attack designs and techniques.
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。
