Detecting Cross-Modal Inconsistency to Defend Against Neural Fake News

Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, pages 2081–2106, November 16–20, 2020. c⃝2020 Association for Computational Linguistics 2081 Detecting Cross-Modal Inconsistency to Defend Against Neural Fake News Reuben Tan Boston University rxtan@bu.edu Bryan A. Plummer Boston University bplum@bu.edu Kate Saenko Boston University MIT-IBM Watson AI Lab saenko@bu.edu Abstract Large-scale dissemination of disinformation online intended to mislead or deceive the gen- eral population is a major societal problem. Rapid progression in image, video, and natural language generative models has only exacer- bated this situation and intensified our need for an effective defense mechanism. While exist- ing approaches have been proposed to defend against neural fake news, they are generally constrained to the very limited setting where articles only have text and metadata such as the title and authors. In this paper, we introduce the more realistic and challenging task of de- fending against machine-generated news that also includes images and captions. To iden- tify the possible weaknesses that adversaries can exploit, we create a NeuralNews dataset composed of 4 different types of generated ar- ticles as well as conduct a series of human user study experiments based on this dataset. In addition to the valuable insights gleaned from our user study, we provide a relatively effective approach based on detecting visual- semantic inconsistencies, which will serve as an effective first line of defense and a useful reference for future work in defending against machine-generated disinformation. Our code and dataset can be downloaded from here. 1 Introduction The rapid progression of generative models in both computer vision (Goodfellow et al., 2014; Zhang et al., 2017, 2018; Choi et al., 2018) and natural language processing (Jozefowicz et al., 2016; Rad- ford et al., 2018, 2019) has led to the increasing likelihood of realistic-looking news articles gener- ated by Artificial Intelligence (AI). The malicious use of such technology could present a major so- cietal problem. Zellers et al. (2019) report that humans are easily deceived by its AI-generated propaganda. By manipulating such technology, Parliament was scheduled to reconvene on Oct 9, but Mr. Johnson said he plan- ned to extend its break. nytimes.com What's Next for Britons after Brexit? August 28, 2019 - Anne Smith In September, voters overwhelming rejected a plan from Prime Minister Theresa May’s team for the United Kingdom to stay in the European Union. On March 29, Britain will officially exit the union after years of campaigning and serious negotiations. The EU’s chief Brexit negotiator, Michel Barnier, has warned that there could be no future trade deals with the United Kingdom if there is a “no deal.” The transition period will allow the United Kingdom and the European Union to work out a new plan for their relationship. But we may not know ... article photo caption model Human or machine-generated? Figure 1: We propose an approach for detecting news articles generated by machines. Prior work uses only the article con- tent and metadata including title, date, domain, and authors. However, news articles often contain …

Content truncated to 500 words