English

Findings of Factify 2: Multimodal Fake News Detection

Computation and Language 2023-09-14 v2 Computer Vision and Pattern Recognition

Abstract

With social media usage growing exponentially in the past few years, fake news has also become extremely prevalent. The detrimental impact of fake news emphasizes the need for research focused on automating the detection of false information and verifying its accuracy. In this work, we present the outcome of the Factify 2 shared task, which provides a multi-modal fact verification and satire news dataset, as part of the DeFactify 2 workshop at AAAI'23. The data calls for a comparison based approach to the task by pairing social media claims with supporting documents, with both text and image, divided into 5 classes based on multi-modal relations. In the second iteration of this task we had over 60 participants and 9 final test-set submissions. The best performances came from the use of DeBERTa for text and Swinv2 and CLIP for image. The highest F1 score averaged for all five classes was 81.82%.

Keywords

Cite

@article{arxiv.2307.10475,
  title  = {Findings of Factify 2: Multimodal Fake News Detection},
  author = {S Suryavardan and Shreyash Mishra and Megha Chakraborty and Parth Patwa and Anku Rani and Aman Chadha and Aishwarya Reganti and Amitava Das and Amit Sheth and Manoj Chinnakotla and Asif Ekbal and Srijan Kumar},
  journal= {arXiv preprint arXiv:2307.10475},
  year   = {2023}
}

Comments

Defactify2 @AAAI 2023