Cyberbullying is a growing problem across social media platforms, inflicting short and long-
lasting effects on victims. To mitigate this problem, research has looked into building …

Large language models (LLMs) are increasingly deployed as the backend for various
applications, including code completion tools and AI-powered search engines. Unlike …

Language models (LMs) are becoming the foundation for almost all major language
technologies, but their capabilities, limitations, and risks are not well understood. We present …

Recent advances in instruction-following large language models (LLMs) have led to
dramatic improvements in a range of NLP tasks. Unfortunately, we find that the same …

We present a holistic approach to building a robust and useful natural language
classification system for real-world content moderation. The success of such a system relies …

Growing concern surrounds the impact of social media platforms on public discourse,,–and
their influence on social dynamics,,,–, especially in the context of toxicity,–. Here, to better …

Abstract⚠ This paper contains prompts and model outputs that are offensive in nature. When
trained on large, unfiltered crawls from the Internet, language models pick up and reproduce …

Despite remarkable advances that large language models have achieved in chatbots,
maintaining a non-toxic user-AI interactive environment has become increasingly critical …

The rapid advancement of large language models (LLMs) has revolutionized artificial
intelligence, introducing unprecedented capabilities in natural language processing and …

Abstract Machine learning approaches often require training and evaluation datasets with a
clear separation between positive and negative examples. This requirement overly …