Target Span Detection for Implicit Harmful Content
This work addresses the challenge of detecting implicit hate speech on online platforms, which is important for protecting vulnerable groups, but it is incremental as it builds on existing datasets and tasks.
The paper tackled the problem of identifying implied targets in hate speech, which is crucial for detecting subtler harmful content online, by creating a new dataset called Implicit-Target-Span from three existing datasets using a pooling method with human annotations and LLMs, and found it provides a challenging test bed for detection methods.
Identifying the targets of hate speech is a crucial step in grasping the nature of such speech and, ultimately, in improving the detection of offensive posts on online forums. Much harmful content on online platforms uses implicit language especially when targeting vulnerable and protected groups such as using stereotypical characteristics instead of explicit target names, making it harder to detect and mitigate the language. In this study, we focus on identifying implied targets of hate speech, essential for recognizing subtler hate speech and enhancing the detection of harmful content on digital platforms. We define a new task aimed at identifying the targets even when they are not explicitly stated. To address that task, we collect and annotate target spans in three prominent implicit hate speech datasets: SBIC, DynaHate, and IHC. We call the resulting merged collection Implicit-Target-Span. The collection is achieved using an innovative pooling method with matching scores based on human annotations and Large Language Models (LLMs). Our experiments indicate that Implicit-Target-Span provides a challenging test bed for target span detection methods.