The fields of humanoid robotics, motion planning, and multimodal intelligence are experiencing significant growth, driven by advancements in open-source platforms, optimization techniques, and large language models. In humanoid robotics, researchers are exploring new methods for designing and controlling robots, including sheet metal welding and physics-based control. Noteworthy papers include MEVITA, HuBE, and FARM, which demonstrate robust walking behaviors, human-like behaviors, and improved tracking accuracy. In motion planning, innovative optimization techniques such as direction-informed and force-directed approaches are enhancing the convergence rate and solution quality of algorithms. Papers like DIT*, APT*, and GIT* propose novel sampling-based planners that improve the efficiency and effectiveness of motion planning. The field of multimodal intelligence is moving towards more embodied and spatially-aware approaches, with researchers exploring the role of morphology and spatial cognition in intelligent behavior. Papers like Morphological Cognition, MaRVL-QA, and 11Plus-Bench demonstrate the potential of multimodal large language models for mathematical reasoning, spatial reasoning, and cognitive-inspired analysis. Overall, these advancements are paving the way for more sophisticated and reliable autonomous systems, with potential applications in various fields.