The field of robotics is moving towards more advanced and generalized mobile manipulation capabilities, with a focus on enabling robots to assist humans in dynamic environments. Recent developments have led to the creation of more efficient and effective methods for object reconstruction, grasping, and manipulation. The use of vision-language models and large language models has improved the ability of robots to understand and interact with their environment, and to perform tasks such as object-to-object affordance grounding and robotic manipulation. Notable papers in this area include MoTo, which proposes a plug-in module for empowering manipulation foundation models with mobile manipulation ability, and O$^3$Afford, which presents a one-shot 3D object-to-object affordance learning approach for robotic manipulation. Other noteworthy papers include Cooperative Grasping for Collective Object Transport, which proposes a novel framework for decision-making in cooperative grasping, and ZLATTE, which presents a geometry-aware, learning-free framework for language-driven trajectory reshaping in human-robot interaction.