Short Form Video

Short-form video research centers on understanding and predicting user engagement with these ubiquitous clips, encompassing analysis of visual, audio, and textual content. Current efforts leverage large language models (LLMs) and deep learning architectures like transformers and diffusion models for tasks such as content analysis, engagement prediction, and even video generation. This research is crucial for improving social media algorithms, mitigating the spread of misinformation (especially concerning mental health), and enhancing video quality assessment and generation techniques. The development of large, annotated datasets is a key component of this rapidly evolving field.