arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2606.25152cs.CLcs.AI
收藏

击中移动目标:持续分布漂移下AI文本检测的测试时适应

Hitting a Moving Target: Test-Time Adaptation for AI Text Detection under Continual Distribution Shift

Cornell Tech(康奈尔科技);MIT(麻省理工学院)

Kevin Ren, Manish Raghavan, Nikhil Garg

AI总结:

提出一种基于半监督学习的测试时适应方法,利用推理时未标记样本的同质性,应对AI文本检测中持续出现的分布漂移,显著优于现有监督检测器。

摘要:

部署的AI文本检测方法通常依赖于训练时获取人类撰写和AI生成文本的标记数据集。这种方法容易受到部署后持续发生的三种分布漂移的影响,且标记数据通常不可用:对抗性人性化、新LLM的发布以及人类写作的时间漂移。同时,现有方法没有利用LLM使用的一个关键信号:推理时的同质性。我们提出了一种基于半监督学习的测试时适应(TTA)方法,通过利用推理时观察到的未标记样本的同质性来适应分布漂移。实验发现,最先进的监督检测器在遇到AI生成和人类写作的分布漂移(包括对抗性和自然漂移)时系统性地失败,而基于半监督学习的测试时适应在很大程度上具有鲁棒性;例如,商业模型Pangram仅检测出24.1%的对抗性AI生成文本,而我们的测试时方法检测出90.5%。我们证明测试时适应是野外AI文本检测的一个有前景的框架。我们在该https URL上公开发布了我们的代码(包括模型训练、评估和绘图的代码)。

英文摘要:

Deployed AI text detection approaches are supervised, relying on training-time access to labeled datasets of both human-written and AI-generated text. This approach is vulnerable to three types of distribution shifts that occur continually post-deployment, and for which labeled data is often unavailable: adversarial humanization, new LLMs being released, and unidentifiable temporal drift in human writing after widespread LLM usage. Simultaneously, existing approaches do not leverage a key signal of LLM usage: inference-time homogeneity. We propose a test-time adaptation (TTA) approach that adapts to distribution shifts by leveraging homogeneity among unlabeled samples observed at inference time. Empirically, we find that state-of-the-art supervised detectors systematically fail when they encounter distribution shifts in AI-generated and human writing, both adversarial and natural, while TTA with semi-supervised learning is largely robust; e.g., the commercial model Pangram detects just 27.0% of our adversarial AI-generated text, compared to 93.3% for our test-time approach. We establish that TTA is a promising framework for AI text detection in the wild.

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。

↑