arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2406.04906cs.CVcs.AI
收藏

RU-AI:用于机器生成内容检测的大型多模态数据集

RU-AI: A Large Multimodal Dataset for Machine-Generated Content Detection

University of Technology Sydney(悉尼科技大学);University of New South Wales(新南威尔士大学);Macquarie University(麦考瑞大学);The University of Sydney(悉尼大学)

Liting Huang, Zhihao Zhang, Yiran Zhang, Xiyue Zhou, Shoujin Wang

AI总结:

针对机器生成内容检测缺乏对齐多模态数据集的问题,本文构建包含147万余实例的三模态数据集RU-AI并创建噪声变体,经实验发现现有SOTA模型检测性能不足,该数据集可推动相关研究与生成式AI的负责任使用。

摘要:

近年来,生成式AI模型创建逼真且类人内容的能力正在显著改变人们交流、创作和工作的方式。机器生成内容是一把双刃剑:一方面,合理使用可造福社会;另一方面,它可能误导公众,对社会构成威胁,尤其是与人类创作的自然内容混合时。因此,迫切需要开发有效方法检测机器生成内容。然而,缺乏对齐的多模态数据集阻碍了此类方法的发展,特别是在文本、图像、语音三模态场景中。本文介绍RU-AI,一个用于文本、图像和语音中机器生成内容鲁棒有效检测的大型多模态数据集。该数据集基于Flickr8K、COCO和Places205三个大型公开数据集构建,通过添加对应的AI生成副本,共包含1,475,370个实例。此外,我们创建了数据集的额外噪声变体,用于测试检测模型的鲁棒性。我们使用当前最先进(SOTA)的检测方法在该数据集上进行了广泛实验,结果显示现有模型在该数据集上仍难以实现准确且鲁棒的检测。我们希望这个新数据集能推动机器生成内容检测领域的研究,促进生成式AI的负责任使用。源代码和数据集可在https://github.com/ZhihaoZhang97/RU-AI获取。

英文摘要:

The recent generative AI models' capability of creating realistic and human-like content is significantly transforming the ways in which people communicate, create and work. The machine-generated content is a double-edged sword. On one hand, it can benefit the society when used appropriately. On the other hand, it may mislead people, posing threats to the society, especially when mixed together with natural content created by humans. Hence, there is an urgent need to develop effective methods to detect machine-generated content. However, the lack of aligned multimodal datasets inhibited the development of such methods, particularly in triple-modality settings (e.g., text, image, and voice). In this paper, we introduce RU-AI, a new large-scale multimodal dataset for robust and effective detection of machine-generated content in text, image and voice. Our dataset is constructed on the basis of three large publicly available datasets: Flickr8K, COCO and Places205, by adding their corresponding AI duplicates, resulting in a total of 1,475,370 instances. In addition, we created an additional noise variant of the dataset for testing the robustness of detection models. We conducted extensive experiments with the current SOTA detection methods on our dataset. The results reveal that existing models still struggle to achieve accurate and robust detection on our dataset. We hope that this new data set can promote research in the field of machine-generated content detection, fostering the responsible use of generative AI. The source code and datasets are available at https://github.com/ZhihaoZhang97/RU-AI.

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。

↑