The recent developments in the field of recommender systems have shown a significant shift towards integrating more sophisticated techniques to address the challenges of data sparsity and user interaction complexity. A notable trend is the adoption of recursive frameworks that leverage both query and item information to enhance recommendation efficiency, particularly in multiple-round interactions. Additionally, there is a growing emphasis on sequential recommendation systems that utilize future data and enduring hard negatives to improve prediction accuracy, showcasing state-of-the-art performance across various benchmarks. Transformer-based models are also being enhanced with frequency information to better handle next-basket recommendation tasks, demonstrating substantial improvements in recall metrics. Overall, the field is progressing towards more integrated, data-driven, and context-aware recommendation strategies.