COSMOS: Catching Out-of-Context Misinformation with Self-Supervised Learning

Aneja, Shivangi; Bregler, Christoph; Nießner, Matthias

doi:10.48550/arxiv.2101.06278

Cited by 12 publications

(47 citation statements)

References 32 publications

(64 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Aneja et al [5] have created COSMOS, a large-scale dataset of around 200𝐾 images which have been matched with 450𝐾 textual captions from different news websites, blogs, and social media posts. Figure 3 presents the category distribution of the images in the dataset, which were collected from a wide-variety of articles with special focus on topics where misinformation spread is prominent.…”

Section: Datasetmentioning

confidence: 99%

“…• caption_modified: Modified caption after applying Spacy NER 3 . Authors in [5] use this caption as an input to their model during experiments. • entity_list: List of mappings between the modified named entities in caption with the corresponding hypernyms.…”

Section: Training and Validation Splitsmentioning

confidence: 99%

“…2 Prospective participants are able to get access to the dataset by filling out a Google Form indicated on the challenge webpage. Note that for detecting bounding boxes, the authors in [5] use the Detectron2 pretrained model 4 available under 5 . They detect up to 10 bounding boxes per image.…”

Section: Training and Validation Splitsmentioning

confidence: 99%

“…Re-contextualization or out-ofcontext (OOC) misuse, which include associating falsified or unrelated claims with a genuine image in order to misrepresent events or persons is, however, relatively niche and unexplored. Very recently, Aneja et al [5] introduced this task, provided a dataset of real-world news posts called COSMOS, and proposed a method for detecting cheapfakes which was benchmarked using this dataset.…”

Section: Introductionmentioning

confidence: 99%

“…Image sources: [4,[13][14][15]25] idea is to take an existing genuine image and create a highly convincing but potentially misleading message in order to spread misinformation. The challenge is based on the recently compiled COSMOS dataset [5], tailored specifically to the detection of OOC misuse of images. The algorithms developed by participants for this challenge should be able to detect OOC <Image, Caption1, Caption2> triplets in news items.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

MMSys'21 Grand Challenge on Detecting Cheapfakes

Aneja,

Midoglu,

Dang-Nguyen

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Section: Datasetmentioning

confidence: 99%

Section: Training and Validation Splitsmentioning

confidence: 99%

Section: Training and Validation Splitsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

MMSys'21 Grand Challenge on Detecting Cheapfakes

Aneja,

Midoglu,

Dang-Nguyen

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Visual User-Generated Content Verification in Journalism: An Overview

et al. 2023

View full text Add to dashboard Cite

Over the past few years, social media has become an indispensable part of the news generation and dissemination cycle on the global stage. These digital channels along with the easy-to-use editing tools have unfortunately created a medium for spreading mis-/disinformation containing visual content. Media practitioners and fact-checkers continue to struggle with scrutinising and debunking visual user-generated content (UGC) quickly and thoroughly as verification of visual content requires a high level of expertise and could be exceedingly complex amid the existing computational tools employed in newsrooms. The aim of this study is to present a forward-looking perspective on how visual UGC verification in journalism can be transformed by multimedia forensics research. We elaborate on a comprehensive overview of the five elements of the UGC verification and propose multimedia forensics as the sixth element. In addition, different types of visual content forgeries and detection approaches proposed by the computer science research community are explained. Finally, a mapping of the available verification tools media practitioners rely on is created along with their limitations and future research directions to gain the confidence of media professionals in using multimedia forensics tools in their day-to-day routine.

show abstract

The Battlefront of Combating Misinformation and Coping with Media Bias

Fung

Huang

Nakov

et al. 2022

Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

View full text Add to dashboard Cite

Misinformation is a pressing issue in modern society. It arouses a mixture of anger, distrust, confusion, and anxiety that cause damage on our daily life judgments and public policy decisions. While recent studies have explored various fake news detection and media bias detection techniques in attempts to tackle the problem, there remain many ongoing challenges yet to be addressed, as can be witnessed from the plethora of untrue and harmful content present during the COVID-19 pandemic and the international crises of late. In this tutorial, we provide researchers and practitioners with a systematic overview of the frontier in fighting misinformation. Specifically, we dive into the important research questions of how to (i) develop a robust fake news detection system, which not only fact-check information pieces provable by background knowledge but also reason about the consistency and the reliability of subtle details for emerging events; (ii) uncover the bias and agenda of news sources to better characterize misinformation; as well as (iii) correct false information and mitigate news bias, while allowing diverse opinions to be expressed. Moreover, we discuss the remaining challenges, future research directions, and exciting opportunities to help make this world a better place, with safer and more harmonic information sharing.

show abstract

COSMOS: Catching Out-of-Context Misinformation with Self-Supervised Learning

Cited by 12 publications

References 32 publications

MMSys'21 Grand Challenge on Detecting Cheapfakes

MMSys'21 Grand Challenge on Detecting Cheapfakes

Visual User-Generated Content Verification in Journalism: An Overview

The Battlefront of Combating Misinformation and Coping with Media Bias

Contact Info

Product

Resources

About