Search

Social media environments enable users to proliferate misinformation surrounding COVID-19. Expert sources, such as Dr. Anthony Fauci have leveraged social media to present corrective multimedia messages. However, little is known about the efficacy of these messages in revising common misconceptions about COVID-19 and influencing behavior. In this study, we examined the efficacy of a multimodal intervention using authentic social media messages that directly addressed common COVID-19 misconceptions. Going further, we identified individual differences that influenced the effectiveness of the intervention, as well as whether those factors predicted individuals' willingness to receive a COVID-19 vaccine. The results showed that the intervention was successful in increasing knowledge when compared to a baseline control. Those who were older and reported less vaccine hesitancy showed greater learning from the intervention. Factors that significantly predicted intention to vaccinate included receiving the intervention, vaccine confidence, vaccine hesitancy, prior flu vaccination history, age, and fear of COVID-19. These findings indicate that multimodal messages can be effectively leveraged in social media to both fight misinformation and increase intention to be vaccinated - however, these interventions may not be as effective for all individuals.

The ability to automatically assess the quality of paraphrases can be very useful for facilitating literacy skills and providing timely feedback to learners. Our aim is twofold: a) to automatically evaluate the quality of paraphrases across four dimensions: lexical similarity, syntactic similarity, semantic similarity and paraphrase quality, and b) to assess how well models trained for this task generalize. The task is modeled as a classification problem and three different methods are explored: a) manual feature extraction combined with an Extra Trees model, b) GloVe embeddings and a Siamese neural network, and c) using a pretrained BERT model fine-tuned on our task. Starting from a dataset of 1998 paraphrases from the User Language Paraphrase Corpus (ULPC), we explore how the three models trained on the ULPC dataset generalize when applied on a separate, small paraphrase corpus based on children inputs. The best out-of-the-box generalization performance is obtained by the Extra Trees model with at least 75% average F1-scores for the three similarity dimensions. We also show that the Siamese neural network and BERT models can obtain an improvement of at least 5% after fine-tuning across all dimensions.