The field of Embodied AI is moving towards more comprehensive and unified evaluation platforms, enabling systematic understanding of core capabilities and standardized evaluation systems. This direction is driven by the need for scalable and automated acquisition methods for embodied data, as well as the importance of personalized and private robot policies. Noteworthy papers in this area include Embodied Arena, which establishes a systematic embodied capability taxonomy and introduces a standardized evaluation system, and PRoP, which develops a model-agnostic framework for personalized and private robot policies. Additionally, the PersONAL benchmark is introduced, designed to study personalization in Embodied AI and evaluate agents' ability to perceive, reason, and memorize over personalized information.