Researchers at Chalmers University of Technology in Sweden have developed an AI scientist capable of generating scientific hypothesis, designing experiments, and interpreting results. Making new biological discoveries with minimal human intervention is an important advance in self-driving laboratories.Researchers at Chalmers University of Technology in Sweden have developed an AI scientist capable of generating scientific hypothesis, designing experiments, and interpreting results. Making new biological discoveries with minimal human intervention is an important advance in self-driving laboratories.

From sloths to birds, vertebrates rely on insects for a wide array of purposes, including food, shelter, and medicine, according to an international paper co-authored by faculty at Binghamton University, State University of New York. 

Dinos might have been beachcombers that dined on washed-up seafood, a new study has found. Marine plants have higher carbon isotope values than land plants and these reflect in the bodies of the animals that eat them. Researchers showed that carbon isotope traces in fossilized teeth and bones from coastal assemblages were consistently higher than the carbon isotope values of fossils from land-locked assemblages – a signature that could be linked to the consumption of seafood. The authors said the study highlights the connections terrestrial and marine ecosystems have and have had for millions of years.

The research team from National University of Defense Technology systematically described the memory efficiency issues in the training process of large language models (LLMs) with Transformer architecture for scientific fields. As the scale of models continues to expand, memory requirements have become a key bottleneck restricting their applications in scientific fields such as biology, medicine, chemistry, and meteorology. The article systematically reviewed memory-efficient pre-training technologies at the algorithm level, system level, and software-hardware collaborative optimization, and combined with the AlphaFold 2 case, demonstrated how customized memory optimization methods can reduce storage overhead while maintaining prediction accuracy.