The field of natural language processing is moving towards leveraging large language models (LLMs) to analyze and understand social media discourse. Researchers are exploring the capabilities of LLMs to generate effective counter-arguments, improve topic modeling, and detect biases in stance detection tasks. The use of LLMs is enabling the development of more accurate and robust models for analyzing social media data, which can be applied to various domains such as public health, politics, and marketing. Notable papers in this area include:

• A Generalizable Rhetorical Strategy Annotation Model, which proposes a novel framework for automatically generating and labeling synthetic debate data using LLMs.
• Latent Topic Synthesis, which introduces an end-to-end framework for automatically generating an interpretable topic taxonomy from an unlabeled corpus using LLMs.
• Utilising Large Language Models for Generating Effective Counter Arguments to Anti-Vaccine Tweets, which explores the capabilities of LLMs to generate sound counter-argument rebuttals to vaccine misinformation.
• Improving Topic Modeling of Social Media Short Texts with Rephrasing, which develops a model-agnostic framework that leverages LLMs to rephrase raw tweets into more standardized and formal language prior to topic modeling.
• Are Stereotypes Leading LLMs' Zero-Shot Stance Detection, which investigates the bias of LLMs in stance detection tasks and finds that they exhibit significant stereotypes.