报告人简介：

贾韬，重庆师范大学副校长，入选国家人才计划青年项目，研究方向聚焦复杂网络，利用图挖掘、图表示、网络建模等工具，在推荐系统、网络控制、网络传播预测、网络重构、网络安全等主题上开展研究；同时将相关计算工具和建模方法跨学科地应用在社会计算与脑科学领域。研究成果发表于Nature Human Behavior、Nature Communications、PNAS Nexus等综合性期刊，AAAI、SIGKDD、VLDB、IJCAI、CIKM、TKDE、ESWA等CCF会议和期刊，以及Journal of Informetrics、Translational Psychiatry等跨领域期刊，具有鲜明的交叉学科研究特色。

报告摘要：

Recombinant growth theory highlights the pivotal role of cumulative knowledge in driving innovation. Although interconnected knowledge facilitates smoother dissemination, its connection to scientific disruption remains poorly understood. Leveraging 53.8 million publications spanning six decades from the Web of Science, we examine the bibliography of each paper and quantify the degree to which the references therein are ``independent'' in the sense that they do not cite one another. We observe that papers built on independent knowledge have decreased over time. However, propensity score matching and regression analyses reveal that such papers are associated with greater scientific disruption, as those who cite them are less likely to cite their references. Moreover, a team's preference for independent knowledge amplifies its disruptive potential, regardless of team size, geographic distance, or collaboration freshness. Despite the disruptive nature, papers built on independent knowledge receive fewer citations and delayed recognition. Taken together, these findings fill a critical gap in our fundamental understanding of scientific innovation, revealing a universal law in peer recognition: Knowledge independence breeds disruption at the cost of impact.