English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The spread of online misinformation threatens public health, democracy, andthe broader society. While professional fact-checkers form the first line ofdefense by fact-checking popular false claims, they do not engage directly inconversations with misinformation spreaders. On the other hand, non-expertordinary users act as eyes-on-the-ground who proactively counter misinformation-- recent research has shown that 96% counter-misinformation responses are madeby ordinary users. However, research also found that 2/3 times, these responsesare rude and lack evidence. This work seeks to create a counter-misinformationresponse generation model to empower users to effectively correctmisinformation. This objective is challenging due to the absence of datasetscontaining ground-truth of ideal counter-misinformation responses, and the lackof models that can generate responses backed by communication theories. In thiswork, we create two novel datasets of misinformation and counter-misinformationresponse pairs from in-the-wild social media and crowdsourcing fromcollege-educated students. We annotate the collected data to distinguish poorfrom ideal responses that are factual, polite, and refute misinformation. Wepropose MisinfoCorrect, a reinforcement learning-based framework that learns togenerate counter-misinformation responses for an input misinformation post. Themodel rewards the generator to increase the politeness, factuality, andrefutation attitude while retaining text fluency and relevancy. Quantitativeand qualitative evaluation shows that our model outperforms several baselinesby generating high-quality counter-responses. This work illustrates the promiseof generative text models for social good -- here, to help create a safe andreliable information ecosystem. The code and data is accessible onhttps://github.com/claws-lab/MisinfoCorrect.

Reinforcement Learning-based Counter-Misinformation Response Generation: A Case Study of COVID-19 Vaccine Misinformation