MM Soc

MM-Soc research focuses on evaluating and improving the capabilities of multimodal large language models (MLLMs) to understand and process information from social media platforms, which contain diverse data types like text, images, and videos. Current research emphasizes benchmarking MLLMs on tasks such as misinformation and hate speech detection, often using novel datasets and comparing performance across different model architectures and sizes. This work is significant because it aims to enhance the ability of AI systems to navigate the complexities of social media, impacting areas like content moderation, sentiment analysis, and understanding online social dynamics.