The last decade has witnessed exponential growth in online social platforms like Twitter, LinkedIn, and WeChat. These platforms have become influential channels for people to connect and share information. The “word-of-mouth” effect allows information to spread fast on these platforms, making it crucial to understand the underlying mechanisms driving information diffusion and to quantify its impact. A considerable amount of attention has been dedicated to tackling this issue, aiming to deepen our comprehension and enhance outcomes in areas like advertising. Concurrently, neural networks have experienced a remarkable evolution, leading to the development of numerous deep learning models. In comparison to conventional approaches, deep learning techniques consistently demonstrate greater efficacy and utility. This survey provides a comprehensive review of recent works applying deep learning methods to the popularity prediction problem.