Chemists routinely optimize reactions to maximize the yield of their desired products, but understanding why those reactions work can require laborious experiments that track reactions through time. Researchers at the University of Tokyo have developed a method to extract hidden information on reaction speeds from yield data obtained during reaction optimization using machine learning together with rate equations developed by chemists. Concentration-dependent yield analysis (CYAN) uses machine learning to expand the yield data from reaction optimization experiments, then uses this information to calculate how quickly the different steps of a reaction occur. This gives chemists a tool to understand, improve and design complex high-yield chemical reactions. 

Kyoto, Japan -- While silicon carbide -- SiC -- has long been promoted as the key to developing electronics suitable for extreme environments, the field has been stuck in the lab for some time. Now, a team of researchers focusing on junction field-effect transistors, or JFETs, have succeeded in developing a new SiC transistor structure that can operate at 600°C, as announced recently by Kyoto University.* One of the team's goals after their previous study was to scale up to wafer-level production, so that is what they have now accomplished.

Compared with the previous study, wafer-scale production refers to building much smaller components and involves developing a minuscule but more efficient SiC transistor. Progressing to the wafer-scale is essential for practical implementation, as the fabrication of integrated circuits requires a large number of very small transistors.

However, wafer-scale production is virtually impossible to accomplish in a university laboratory due to equipment constraints. This motivated the team to collaborate with Phenitec Semiconductor Corp, which operates a mass-production fabrication plant, making this next step in their research feasible.

Cameras have long been regarded as tools for capturing what the human eye can see. But that assumption may soon be challenged.

“If we change the way we capture data, the potential of Artificial Intelligence (AI) expands dramatically,” says Associate Professor Hiroyuki Kubo of the Graduate School of Informatics. Dr. Kubo is developing methods to visualize what is normally invisible—such as blood vessels beneath the skin or vehicles hidden in fog—using not specialized equipment, but systems built with ordinary projectors and cameras. His approach could significantly accelerate advances in AI. Beyond imaging, Dr. Kubo is also exploring new possibilities in AI-assisted coloring. By designing systems in which humans and AI collaborate to efficiently color large volumes of animation cels, his research aims to enhance productivity in anime production.

We spoke with Dr. Kubo about his work and his vision for the future as he continues to push the boundaries of what cameras—and AI—can do.

Kyoto, Japan -- Our brains internally represent the outside world through the coordinated activity of billions of neurons. A single neuron responds unreliably: even when the same stimulus is shown repeatedly, its activity varies from trial to trial. Neuron populations can compensate for this variability by representing the same information across many neurons: a strategy known as population coding.

However, neurons do not always fluctuate independently. When many neuronal signals rise and fall together, their shared variability -- known as noise correlation -- can overlap with the pattern of activity that carries information about a stimulus, and may eventually cap the information the population can convey. This possibility puzzled an international team of researchers from Kyoto University, Harvard University, and the University of California, Los Angeles.

"We face a fundamental question: why does the brain have so many neurons if shared fluctuations impose a ceiling on information?" asks S. Amin Moosavi of UCLA.

A research team led by Professor Hitoshi Yamamoto of Rissho University experimentally examined how people evaluate others when the public reputation of the person involved in an interaction conflicts with their own personal experience with that person. The study was published in the international journal Evolution and Human Behavior on September 14, 2026.

Build systems are a fundamental component of modern software development. Dependency-related errors represent a significant portion of build errors in large-scale projects. However, existing tools for detecting these errors struggle to achieve both high performance and accuracy. In a new study, researchers developed mkcheck2, which reduces performance overhead by up to 99.7% while maintaining accuracy through an efficient eBPF-based system call tracing method and incremental analysis, thereby making continuous dependency verification practical.

Following the 2011 the Great East Japan Earthquake and tsunami, freshwater and marine threespine stickleback hybridized in a newly formed wetland in Otsuchi, Iwate, Japan. Nine years of genomic tracking revealed rapid elimination of marine-derived genomic regions within ten generations, identifying key reproductive-isolation loci controlling freshwater survival and migration to the sea. The study also found that widespread genetic incompatibilities contribute to genome-wide elimination, providing rare real-time insight into the genetic mechanisms maintaining species boundaries.

A study of older adults in Japan found that routine health checkup measures were associated with healthcare costs 3–4 years later, revealing nonlinear and sex-specific patterns.

A research team from the Vision and Action Laboratory, Visual Perception and Cognition Laboratory, and Cognitive Neurotechnology Unit in the Department of Computer Science and Engineering at Toyohashi University of Technology, led by Associate Professor Hideki Tamura, has shown that the temporal patterns of pupillary responses differ depending on whether observers evaluate the same object images in terms of glossiness or attractiveness. The research team used images of 60 everyday objects as visual stimuli and measured pupil diameter while participants separately evaluated their perceived glossiness and attractiveness. Exactly the same set of images was used in both tasks. The results showed that higher glossiness ratings were associated with greater pupil constriction during a relatively early time window centered at approximately 1 second after stimulus onset. In contrast, higher attractiveness ratings tended to be associated with greater pupil dilation at later time points. These findings suggest that even when viewing the same objects, the temporal characteristics of pupillary responses may differ depending on what observers are evaluating.

The study was published in Journal of Vision on September 1, 2026.

https://doi.org/10.1167/jov.26.9.1