朝向通用人工智能的进展测量：一种认知框架

A framework has been introduced to quantitatively measure progress toward Artificial General Intelligence (AGI), highlighting the significance of cognitive abilities in AGI development. By establishing a systematic evaluation mechanism, researchers can better understand and assess the performance of various AI systems in cognitive tasks. Additionally, to foster the development of relevant evaluation tools, DeepMind has launched a Kaggle hackathon encouraging participants to design and implement effective assessment methods. This initiative not only advances AGI research but also provides a collaborative and innovative platform for the AI community.