The field of robotic grasping and manipulation is experiencing significant developments, with a focus on improving the efficiency, adaptability, and robustness of robotic systems. Researchers are exploring novel approaches to tackle the challenges of uncertain environments, pose uncertainty, and generalize grasping skills to unseen object categories. Notably, new methods are being introduced to guarantee reachability and feasibility of interception under uncertain conditions, as well as to synthesize high-quality dexterous hand configurations for diverse task instructions. Some papers are particularly noteworthy, including: G-DexGrasp, which proposes a retrieval-augmented generation approach for generalizable dexterous grasping synthesis. Learning Adaptive Dexterous Grasping from Single Demonstrations, which introduces a framework that learns a library of grasping skills from single human demonstrations and selects the most suitable one using a vision-language model.