The generation of text that is increasingly human-like has been enabled by
the advent of large language models (LLMs). As the detection of AI-generated
content holds significant importance in the fight against issues such as
misinformation and academic cheating, numerous studies have b
在本文中,我们提出了一个针对更广泛的对抗性攻击类别的框架,旨在对机器生成的内容进行微小扰动以逃避检测,通过对动态场景中的对抗性学习来评估当前检测模型对这种攻击的鲁棒性的潜力提升。实证结果发现,现有的检测模型可以在仅 10 秒内受到破坏,将机器生成的文本误分类为人类撰写的内容。尽管观察到模型鲁棒性方面的一些改进,但实际应用仍面临重大挑战,这些发现为 AI 文本检测器的未来发展提供了启示,强调了对更准确和鲁棒的检测方法的需求。