arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2210.09421cs.CRcs.CLcs.LG
收藏

Deepfake文本检测:限制和机会

Deepfake Text Detection: Limitations and Opportunities

Virginia Tech(弗吉尼亚理工大学);University of Chicago(芝加哥大学);LUMS Pakistan(巴基斯坦拉合尔管理科学大学);University of Virginia(弗吉尼亚大学)

Jiameng Pu, Zain Sarwar, Sifat Muhammad Abdullah, Abdullah Rehman, Yoonjin Kim, Parantapa Bhattacharya, Mobin Javed, Bimal Viswanath

摘要:

语言生成模型的最新进展使得能够创建令人信服的合成文本或deepfake文本。之前的工作已经证明了滥用deepfake文本误导内容消费者的可能性。因此,区分人类和机器生成文本的深度伪造文本检测变得越来越重要。已经提出了几种用于deepfake文本检测的防御方法。然而,我们对它们在现实世界中的适用性缺乏深入的了解。在本文中,我们从基于transformer的工具提供支持的4个在线服务中收集了deepfake文本,以评估防御对野生内容的泛化能力。我们开发了几种低成本的对抗攻击,并研究了现有防御对自适应攻击的鲁棒性。我们发现,许多防御显示出显着的性能下降,在我们的评估方案相比,他们原来声称的性能。我们的评估表明,利用文本内容中的语义信息是提高deepfake文本检测方案的鲁棒性和泛化性能的一种很有前途的方法。

英文摘要:

Recent advances in generative models for language have enabled the creation of convincing synthetic text or deepfake text. Prior work has demonstrated the potential for misuse of deepfake text to mislead content consumers. Therefore, deepfake text detection, the task of discriminating between human and machine-generated text, is becoming increasingly critical. Several defenses have been proposed for deepfake text detection. However, we lack a thorough understanding of their real-world applicability. In this paper, we collect deepfake text from 4 online services powered by Transformer-based tools to evaluate the generalization ability of the defenses on content in the wild. We develop several low-cost adversarial attacks, and investigate the robustness of existing defenses against an adaptive attacker. We find that many defenses show significant degradation in performance under our evaluation scenarios compared to their original claimed performance. Our evaluation shows that tapping into the semantic information in the text content is a promising approach for improving the robustness and generalization performance of deepfake text detection schemes.

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。

↑