Rural and primarily agricultural economies are the most susceptible to climate change and armed conflict. Climate-driven weather events add pressure to an already fragile system, affecting the socioeconomic conditions on which health depends, especially in ethnically divided or areas with scarce resources.

Nowadays, compute-intensive programs, like those for training artificial intelligence and machine learning models, are used extensively. Modern compilers use vectorization techniques to exploit parallel processing capabilities to improve the performance of such programs. A group of scientists from the University of Southern California, Cisco AI Research, and Intel Labs designed a data-driven, graph-based learning framework for automatic vectorization called autograph, which utilizes deep reinforcement learning to have an intelligent agent learn an optimal policy. Autograph greatly outperformed other approaches across different datasets. The work was published on June 2, 2025, in an article titled “A Graph-Based Learning Framework for Compiler Loop Auto-Vectorization” in Intelligent Computing, a Science Partner Journal.

Land use is at the heart of the many emergencies facing our world today: climate change, biodiversity loss, social injustice and food insecurity. These – exacerbated by unsustainable practices such as industrial agriculture – combine to create simultaneous and interconnected crises. To overcome these challenges, large parts of the Earth's surface must meet several demands at the same time – from species conservation and food production to human well-being. This is where multifunctional landscapes come in, because they can meet many ecological, social and economic goals at once. Researchers at the Universities of Göttingen and Kassel have evaluated how using land in such ways can support nature conservation and ecosystem restoration. Their review article shows ways to redesign land use and nature conservation using integrated approaches. This was published in Nature Reviews Biodiversity.

The University of Copenhagen today launches a new AI Package aimed at helping Denmark and Europe harness the potential of artificial intelligence in a responsible and sustainable way — for the benefit of many.

Dynamic multi-robot task allocation (MRTA) requires real-time responsiveness and adaptability to rapidly changing con ditions. Existing methods, primarily based on static data and centralized architectures, often fail in dynamic environments that require decentralized, context-aware decisions. To address these challenges, this paper proposes a novel graph reinforce ment learning (GRL) architecture, named Spatial-Temporal Fusing Reinforcement Learning (STFRL), to address real-time distributed target allocation problems in search and rescue scenarios. The proposed policy network includes an encoder, which employs a Temporal-Spatial Fusing Encoder (TSFE) to extract input features and a decoder uses multi-head attention (MHA) to perform distributed allocation based on the encoder’s output and context. The policy network is trained with the REINFORCEalgorithm.Experimentalcomparisonswithstate-of-the-artbaselinesdemonstratethatSTFRLachievessuperior performance in path cost, inference speed, and scalability, highlighting its robustness and efficiency in complex, dynamic environments.

A research team led by Professor Yumei Zhang from the Academy of Global Food Economics and Policy (AGFEP) at China Agricultural University, in collaboration with Dr. Issa Ouedraogo from the Alliance of Biodiversity International and International Center for Tropical Agriculture (CIAT), has conducted a study that reveals the structural characteristics and emission reduction potential of agrifood system carbon emissions in China and Africa. The related article has been published in Frontiers of Agricultural Science and Engineering (DOI: 10.15302/J-FASE-2025609).