Science progresses by building upon previous discoveries. It is commonly believed that the impact of scientific papers, as measured by citations, is positively correlated with the impact of past discoveries built upon. However, analyzing over 30 million papers and nearly a billion citations across multiple disciplines, we find that there is a long-term positive citation correlation, but a negative short-term correlation. We demonstrate that the key to resolving this paradox lies in a new concept, called "capacity", which captures the amount of originality remaining for a paper. We find there is an intimate link between capacity and impact dynamics that appears universal across the diverse fields we studied. The uncovered capacity measure not only explains the correlated impact dynamics across the sciences but also improves our understanding and predictions of high-impact discoveries.

翻译：科学通过建立在先前发现的基础上而进步。人们普遍认为，以引用次数衡量的科学论文的影响力，与其所基于的过去发现的影响力正相关。然而，通过分析跨多个学科的超过3000万篇论文和近10亿条引用记录，我们发现存在长期的正向引用相关性，但存在短期负相关。我们证明，解决这一悖论的关键在于一个新概念——“容量”，它捕捉了一篇论文中剩余原创性的数量。我们发现，在我们研究的各个不同领域中，容量与影响力动态之间存在一种普遍存在的密切联系。所揭示的容量度量不仅解释了整个科学领域中的相关性影响动态，还提高了我们对高影响力发现的理解和预测能力。