AI生成的文本检测器对对抗性扰动是否具有鲁棒性?
Are AI-Generated Text Detectors Robust to Adversarial Perturbations?
AI总结:
针对AI生成文本检测器易受对抗扰动影响的问题,提出孪生校准重建网络(SCRN),通过加噪去噪提取鲁棒语义表示并校准置信度,在四个数据集上显著提升检测准确率与泛化能力。
摘要:
大型语言模型(LLM)的广泛使用引发了人们对AI生成文本可能被滥用的担忧,因为这些模型能够生成与人类撰写的文本极为相似的内容。当前的AI生成文本(AIGT)检测器缺乏对抗性扰动的鲁棒性,即使是字符或单词的微小变化也会导致区分人类创作文本与AI生成文本的结果发生逆转。本文研究了现有AIGT检测方法的鲁棒性,并提出了一种新型检测器——孪生校准重建网络(SCRN)。SCRN采用重建网络对文本进行加噪和去噪,提取对局部扰动具有鲁棒性的语义表示。我们还提出了一种孪生校准技术,训练模型在不同噪声下做出同等置信度的预测,从而提升模型对对抗性扰动的鲁棒性。在四个公开数据集上的实验表明,SCRN在所有基线方法中表现最佳,在对抗攻击下相较于最优基线方法取得了6.5%至18.25%的绝对准确率提升。此外,它在跨领域、跨体裁和混合来源场景中展现出优越的泛化能力。代码可在https://github.com/CarlanLark/Robust-AIGC-Detector获取。
英文摘要:
The widespread use of large language models (LLMs) has sparked concerns about the potential misuse of AI-generated text, as these models can produce content that closely resembles human-generated text. Current detectors for AI-generated text (AIGT) lack robustness against adversarial perturbations, with even minor changes in characters or words causing a reversal in distinguishing between human-created and AI-generated text. This paper investigates the robustness of existing AIGT detection methods and introduces a novel detector, the Siamese Calibrated Reconstruction Network (SCRN). The SCRN employs a reconstruction network to add and remove noise from text, extracting a semantic representation that is robust to local perturbations. We also propose a siamese calibration technique to train the model to make equally confidence predictions under different noise, which improves the model's robustness against adversarial perturbations. Experiments on four publicly available datasets show that the SCRN outperforms all baseline methods, achieving 6.5\%-18.25\% absolute accuracy improvement over the best baseline method under adversarial attacks. Moreover, it exhibits superior generalizability in cross-domain, cross-genre, and mixed-source scenarios. The code is available at \url{https://github.com/CarlanLark/Robust-AIGC-Detector}.
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。