Research team debuts the first visual pre-training paradigm tailored for CTR prediction, lifting Taobao GMV by 0.88% (p < 0.01) via user-intent contrastive learning that mines interest-centric image signals.

FPMCO decomposes multi-constraint RL into KL-projection sub-problems, achieving higher reward with lower computing than second-order rivals on the new SCIG robotics benchmark.

Researchers have developed a hybrid material by integrating bipyramid-shaped gold nanocrystals with sharp edge with a covalent organic framework. This novel structure harnesses concentrated light energy at the gold tips, significantly enhancing the efficiency of photocatalytic hydrogen evolution—a clean method to produce fuel from water and sunlight.

Sweet potato is a globally important food crop with rich nutritional value, widely used in food processing, feed production, and the pharmaceutical industry. Hainan Province in China has a history of sweet potato cultivation spanning over 300 years and has long been one of the major sweet potato-producing areas in southern China.

As the global population continues to grow, climate change intensifies, and agricultural resources become increasingly scarce, traditional agricultural production models are facing unprecedented challenges. It is predicted that by 2050, global food production will need to increase by 70% to meet demand. Against this backdrop, “Agriculture 4.0” has gradually emerged, with embedded technologies such as the Internet of Things (IoT), artificial intelligence (AI), unmanned aerial vehicles (UAVs), and robotics widely applied in precision agriculture to improve production efficiency and sustainability.

In the field of modern agricultural machinery, pneumatic conveying seed-discharge systems have been widely adopted due to their compact size, high seed-discharge efficiency, and strong adaptability, which are particularly aligned with the current development trend of high-speed mechanized operations.

Recent survey delivers the first systematic benchmark of TSP solvers spanning end-to-end deep learners, hybrid methods and brand-new LLM-based hybrids, revealing that hybrids give best-in-class routes in record time.

Meet ncRNA2MetS v2.0, the largest experimentally verified repository linking 1,505 ncRNAs to metabolic syndrome and its five core disorders, offered through an intuitive eight-module web portal.

Researchers introduce a group-driven initialization that fuses search history with graph modularity, boosting state-of-the-art local solvers for k-quasi-clique and k-plex without altering their search routines.