Core Insights - The company is making steady progress in the research and development of its second-generation in-memory computing technology IP [2] - The goal is to achieve a significant increase in single-core NPU computing power and a substantial optimization in energy efficiency [2] - The technology will fully support Transformer models, indicating a focus on advanced AI applications [2]

Core Viewpoint - The company is focusing its R&D investments on low-power high-performance computing, ecosystem development, and wireless connectivity to meet the massive chip demand for edge AI applications [1] Group 1: R&D Focus - The company's R&D efforts are directed towards three main areas: low-power high-performance computing, ecosystem development, and wireless connectivity [1] - The second-generation in-memory computing technology IP development is progressing steadily [1] Group 2: Performance Goals - The goal is to achieve a significant increase in single-core NPU computing power and a substantial optimization in energy efficiency for the next generation of chips [1] - The new chips will fully support Transformer models [1]