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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

The exponential growth of social media platforms such as Twitter and Facebookhas revolutionized textual communication and textual content publication inhuman society. However, they have been increasingly exploited to propagatetoxic content, such as hate speech, malicious advertisement, and pornography,which can lead to highly negative impacts (e.g., harmful effects on teen mentalhealth). Researchers and practitioners have been enthusiastically developingand extensively deploying textual content moderation software to address thisproblem. However, we find that malicious users can evade moderation by changingonly a few words in the toxic content. Moreover, modern content moderationsoftware performance against malicious inputs remains underexplored. To thisend, we propose MTTM, a Metamorphic Testing framework for Textual contentModeration software. Specifically, we conduct a pilot study on 2,000 textmessages collected from real users and summarize eleven metamorphic relationsacross three perturbation levels: character, word, and sentence. MTTM employsthese metamorphic relations on toxic textual contents to generate test cases,which are still toxic yet likely to evade moderation. In our evaluation, weemploy MTTM to test three commercial textual content moderation software andtwo state-of-the-art moderation algorithms against three kinds of toxiccontent. The results show that MTTM achieves up to 83.9%, 51%, and 82.5% errorfinding rates (EFR) when testing commercial moderation software provided byGoogle, Baidu, and Huawei, respectively, and it obtains up to 91.2% EFR whentesting the state-of-the-art algorithms from the academy. In addition, weleverage the test cases generated by MTTM to retrain the model we explored,which largely improves model robustness (0% to 5.9% EFR) while maintaining theaccuracy on the original test set.

MTTM: Metamorphic Testing for Textual Content Moderation Software