Stanford AI Laboratory proposes a zero-shot world model, narrowing the gap between AI and human children's visual learning data.

ME News Report, April 15 (UTC+8), Stanford AI Laboratory (StanfordAILab) recently pointed out that the amount of data required for the most advanced AI models to achieve visual capabilities far exceeds that of human children by several orders of magnitude. To bridge this gap, researchers proposed the Zero-shot World Model (ZWM) approach. This method has made significant progress, with the BabyZWM model achieving performance comparable to an unspecified benchmark, despite being trained using only first-person perspective data from a single child. (Source: InFoQ)