In data center networks, the growth rate of switch buffer size is much smaller than that of network bandwidth capacity. This results in frequent network congestion, which requires congestion control. Currently, the classification of end-to-end congestion control protocols is based on single dimensions such as time or congestion arbitration subjects.

A Chinese research team has achieved a breakthrough in improving the training efficiency of Graph Neural Networks (GNNs). They introduced an innovative architecture named "Decentralized Hypercube Collaborative Framework," addressing long-standing challenges such as high memory overhead, low computational efficiency, and underutilized hardware resources in traditional Graph Convolutional Network (GCN) training.

A comprehensive Cal Poly study examining North American bird population data reveals that environmental change does not affect all species equally. The research, assessing data collected across North America over nearly 30 years, found that no single environmental factor explained bird population changes across all species. Instead, species responded in highly individualized ways because of their interactions with their environment. While human-caused disturbances such as light pollution and air pollution can negatively affect bird populations, ecological traits can help predict species' sensitivity and forecast how well they will respond to ongoing environmental change. The study also highlights the importance of local conservation action. While climate change and regional air pollution require broad policy solutions, many environmental factors affecting birds can be addressed at the level of cities or counties.

A research team has proposed a new framework for understanding how traditional Chinese medicine (TCM) works at the tissue level by using artificial intelligence (AI), single-cell omics, and spatial omics to decode “cell niches”—local microenvironments where multiple cell types interact. 

Recently, a team from the University of Electronic Science and Technology of China (in collaboration with the Beijing Institute of Nanoenergy and Nanosystems, Chinese Academy of Sciences) made significant progress in strain-modulated optoelectronic perception and neuromorphic vision systems. Their findings were published in Nano Research on February 26, 2026.