arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2303.01217cs.MMcs.AIcs.CL
收藏

合成误导者:生成和打击多模式错误信息

Synthetic Misinformers: Generating and Combating Multimodal Misinformation

School of Electrical & Computer Engineering, Aristotle University of Thessaloniki(塞萨洛尼基亚里士多德大学电气与计算机工程学院)

Stefanos-Iordanis Papadopoulos, Christos Koutlis, Symeon Papadopoulos, Panagiotis C. Petrantonakis

摘要:

随着社交媒体的扩大和多媒体内容的传播,错误信息的传播已成为一个主要问题。这就需要有效的策略,多模态错误信息检测(MMD),检测图像及其伴随的文本的组合是否会误导或误导。由于深度神经网络的数据密集型性质和人工注释的劳动密集型过程,研究人员一直在探索各种自动生成合成多模态错误信息的方法-我们将其称为合成Misinformers -以训练MMD模型。然而,由于对真实世界误报的评估有限,而且缺乏与其他合成误报的比较,因此难以评估该领域的进展。为了解决这个问题,我们对现有的和新的合成Misinformers进行了比较研究,涉及(1)上下文外(OOC)图像-标题对,(2)跨模态命名实体不一致(NEI)以及(3)混合方法,并使用COSMOS基准对它们进行了评估。比较研究表明,我们提出的基于CLIP的命名实体交换可以导致MMD模型在多模态准确性方面超过其他OOC和NEI Misinformers,并且混合方法可以导致更高的检测准确性。然而,在减轻COSMOS评估协议的信息泄漏后,低敏感性分数表明该任务比以前的研究所建议的更具挑战性。最后,我们的研究结果表明,基于NEI的合成Misinformers往往会受到单峰偏差的影响,其中纯文本MMD可以胜过多模态MMD。

英文摘要:

With the expansion of social media and the increasing dissemination of multimedia content, the spread of misinformation has become a major concern. This necessitates effective strategies for multimodal misinformation detection (MMD) that detect whether the combination of an image and its accompanying text could mislead or misinform. Due to the data-intensive nature of deep neural networks and the labor-intensive process of manual annotation, researchers have been exploring various methods for automatically generating synthetic multimodal misinformation - which we refer to as Synthetic Misinformers - in order to train MMD models. However, limited evaluation on real-world misinformation and a lack of comparisons with other Synthetic Misinformers makes difficult to assess progress in the field. To address this, we perform a comparative study on existing and new Synthetic Misinformers that involves (1) out-of-context (OOC) image-caption pairs, (2) cross-modal named entity inconsistency (NEI) as well as (3) hybrid approaches and we evaluate them against real-world misinformation; using the COSMOS benchmark. The comparative study showed that our proposed CLIP-based Named Entity Swapping can lead to MMD models that surpass other OOC and NEI Misinformers in terms of multimodal accuracy and that hybrid approaches can lead to even higher detection accuracy. Nevertheless, after alleviating information leakage from the COSMOS evaluation protocol, low Sensitivity scores indicate that the task is significantly more challenging than previous studies suggested. Finally, our findings showed that NEI-based Synthetic Misinformers tend to suffer from a unimodal bias, where text-only MMDs can outperform multimodal ones.

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。

↑