Researchers from Japan and Seattle's Allen Institute have created a detailed supercomputer simulation of a mouse cortex, featuring nearly 10 million neurons and 26 billion synapses, using the world's fastest supercomputer Fugaku. This breakthrough could lead to new methods for studying brain diseases like Alzheimer's and epilepsy.
DeepScientist is a goal-oriented, fully autonomous scientific discovery system. It uses Bayesian Optimization and a hierarchical 'hypothesize, verify, and analyze' process with a Findings Memory to balance exploration and exploitation. It generated and validated thousands of scientific ideas, surpassing human SOTA on three AI tasks.
Large language models (LLMs) are rapidly being implemented in a wide range of disciplines, with the promise of unlocking new possibilities for scientific exploration. However, while the development of LLMs brings opportunities to science, it also comes with pressing challenges. This Focus discusses the current state of the art, highlights key obstacles, and examines some of the potential pitfalls and biases of implementing and using LLMs across different domains, including healthcare, urban planning, chemistry, linguistics, humanities, and computer science. In addition, the Focus explores emerging technologies – such as neuromorphic engineering – that show promise in enhancing the energy efficiency of LLM deployment on hardware platforms.
Neuroscience research suggests that scientists may have been undervaluing the most ancient regions of the human brain when studying consciousness. Evidence indicates that the subcortex and cerebellum may play a more significant role than previously thought, and could even be sufficient for basic forms of consciousness.
Scientists have mapped the activity that takes place across a mouse's entire brain as it decides how to complete a task - and the results could explain the origin of our gut feelings.
It is with profound sadness that we announce the passing of Prof. Lucia Vaina, a remarkable woman whose life was a testament to compassion, perseverance, and an unyielding pursuit of knowledge. She left us on July 25, 2025.
The Langlands programme has inspired and befuddled mathematicians for more than 50 years. A major advance has now opened up new worlds for them to explore.
The article details the recent proof of the geometric Langlands conjecture, a significant advancement in mathematics that validates a decades-old program aiming for a "grand unified theory" of the field. Led by Dennis Gaitsgory and Sam Raskin, the proof—spanning five papers and nearly 1,000 pages—is expected to open new avenues of research and potentially bridge connections between mathematics and theoretical physics, particularly in understanding symmetries in quantum field theory. While not a complete solution to the broader Langlands program, it provides strong evidence for its underlying principles and offers new tools for tackling complex mathematical problems.
The article details the recent proof of the geometric Langlands conjecture, a significant advancement in mathematics that validates a decades-old program aiming for a "grand unified theory" of the field. Led by Dennis Gaitsgory and Sam Raskin, the proof—spanning five papers and nearly 1,000 pages—is expected to open new avenues of research and potentially bridge connections between mathematics and theoretical physics, particularly in understanding symmetries in quantum field theory. While not a complete solution to the broader Langlands program, it provides strong evidence for its underlying principles and offers new tools for tackling complex mathematical problems.
An international research team found that cats prefer to sleep on their left side, potentially as an evolutionary advantage for hunting and escape behavior. Sleeping on the left allows them to process threats with the right hemisphere of the brain, which specializes in spatial awareness and rapid movements.
Human high-order thalamic nuclei gate conscious perception through the thalamofrontal loop
Researchers investigated the neural basis of conscious perception using intracranial recordings in humans. They discovered that the intralaminar and medial thalamic nuclei (imTha) exhibit early and strong activity linked to consciousness, acting as a “gate” that drives information to the prefrontal cortex. This supports the idea that the thalamofrontal loop is essential for visual awareness and provides direct evidence for the thalamus’s role in the rapid emergence of conscious experience.
Researchers investigated the neural basis of conscious perception using intracranial recordings in humans. They discovered that the intralaminar and medial thalamic nuclei (imTha) exhibit early and strong activity linked to consciousness, acting as a “gate” that drives information to the prefrontal cortex. This supports the idea that the thalamofrontal loop is essential for visual awareness and provides direct evidence for the thalamus’s role in the rapid emergence of conscious experience.
A new study published in Science has identified the thalamus as a central player in how humans become consciously aware of visual information. Researchers discovered that specific thalamic regions activate earlier and more strongly during moments of visual awareness, suggesting they act as a gateway for conscious perception.
