The field of human-robot collaboration and social interaction is rapidly advancing, with a focus on developing robots that can effectively interact with humans in various scenarios. Recent research has explored the use of control schemes, physics-informed neural networks, and multimodal human-intent modeling to improve human-robot collaboration. These advancements have the potential to enhance the safety and efficiency of human-robot interactions, particularly in areas such as object manipulation, handovers, and navigation. Noteworthy papers in this area include: Physics-informed Neural Time Fields for Prehensile Object Manipulation, which proposes a novel approach to solving object manipulation tasks, and Multimodal Human-Intent Modeling for Contextual Robot-to-Human Handovers of Arbitrary Objects, which presents a unified approach to selecting target objects and performing handovers based on human preferences. Additionally, the Mixed-Initiative Dialog for Human-Robot Collaborative Manipulation paper introduces a system that enables effective collaboration between humans and robots through natural language dialog.