Researchers have found that children's vocabulary growth accelerates over time, contrary to previous models that described it as a steady accumulation of words. In contrast, language models do not show this acceleration and instead process new data at a constant rate. This difference in learning efficiency may be due to the fact that children use significantly less training data than language models. --- Why it matters: This study's findings have implications for understanding how children learn language and how AI systems can be designed to mimic human-like learning abilities. The discovery of accelerating returns in word learning could inform the development of more efficient and effective language learning algorithms. Source: https://arxiv.org/abs/2608.17120