Toggle light / dark theme

MIT researchers developed a new approach for assessing predictions with a spatial dimension, like forecasting weather or mapping air pollution.

Re relying on a weather app to predict next week’s temperature. How do you know you can trust its forecast? Scientists use statistical and physical models to make predictions about everything from weather to air pollution. But checking whether these models are truly reliable is trickier than it seems—especially when the locations where we have validation data don Traditional validation methods struggle with this problem, failing to provide consistent accuracy in real-world scenarios. In this work, researchers introduce a new validation approach designed to improve trust in spatial predictions. They define a key requirement: as more validation data becomes available, the accuracy of the validation method should improve indefinitely. They show that existing methods don’t always meet this standard. Instead, they propose an approach inspired by previous work on handling differences in data distributions (known as “covariate shift”) but adapted for spatial prediction. Their method not only meets their strict validation requirement but also outperforms existing techniques in both simulations and real-world data.

By refining how we validate predictive models, this work helps ensure that critical forecasts—like air pollution levels or extreme weather events—can be trusted with greater confidence.


A new evaluation method assesses the accuracy of spatial prediction techniques, outperforming traditional methods. This could help scientists make better predictions in areas like weather forecasting, climate research, public health, and ecological management.

In today’s AI news, Backed by $200 million in funding, Scott Wu and his team at Cognition are building an AI tool that could potentially disintegrate the whole industry, at a $2 Billion valuation. Devin is an autonomous AI agent that, in theory, writes the code itself—no people involved—and can complete entire projects typically assigned to developers.

In other advancements, OpenAI is changing how it trains AI models to explicitly embrace “intellectual freedom … no matter how challenging or controversial a topic may be,” the company says in a new policy. OpenAI is releasing a significantly expanded version of its Model Spec, a document that defines how its AI models should behave — and is making it free for anyone to use or modify.

Then, xAI, the artificial intelligence company founded by Elon Musk, is set to launch Grok 3 on Monday, Feb. 17. According to xAI, this latest version of its chatbot, which Musk describes as “scary smart,” represents a major step forward, improving reasoning, computational power and adaptability. Grok 3’s development was accelerated by its Colossus supercomputer, which was built in just eight months, powered by 100,000 Nvidia H100 GPUs.

And, large language models can learn complex reasoning tasks without relying on large datasets, according to a new study by researchers at Shanghai Jiao Tong University. Their findings show that with just a small batch of well-curated examples, you can train an LLM for tasks that were thought to require tens of thousands of training instances.

S new o1 model, which focuses on slower, more deliberate reasoning — much like how humans think — in order to solve complex problems. ” + Then, join Turing Award laureate Yann LeCun—Chief AI Scientist at Meta and Professor at NYU—as he discusses with Link Ventures’ John Werner, the future of artificial intelligence and how open-source development is driving innovation. In this wide-ranging conversation, LeCun explains why AI systems won’t “take over” but will instead serve as empowering assistants.

This process, which cannot be understood satisfactorily by classical physics alone, occurs constantly in green plants and other photosynthetic organisms, such as photosynthetic bacteria. However, the exact mechanisms have still not been fully elucidated. Hauer and first author Erika Keil see their study as an important new basis in the effort to clarify how chlorophyll, the pigment in leaf green, works.

Applying these findings in the design of artificial photosynthesis units could help to utilize solar energy with unprecedented efficiency for electricity generation or photochemistry.

Attosecond time-resolved experiments have revealed the increasing importance of electronic correlations in the collective plasmon response as the size of the system decreases to sub-nm scales.

The study, published in the journal Science Advances, was led by the University of Hamburg and DESY as part of a collaboration with Stanford, SLAC National Accelerator Laboratory, Ludwig-Maximilians-Universität München, Northwest Missouri State University, Politecnico di Milano and the Max Planck Institute for the Structure and Dynamics of Matter.

Plasmons are collective electronic excitations that give rise to unique effects in matter. They provide a means of achieving extreme light confinement, enabling groundbreaking applications such as efficient solar energy harvesting, ultrafine sensor technology, and enhanced photocatalysis.

Transparent aluminum oxide (TAlOx), a real material despite its sci-fi name, is incredibly hard and resistant to scratches, making it perfect for protective coatings on electronics, optical sensors, and solar panels. On the sci-fi show Star Trek, it is even used for starship windows and spacefaring aquariums.

Current methods of making TAlOx are expensive and complicated, requiring high-powered lasers, vacuum chambers, or large vats of dangerous acids. That may change thanks to research co-authored by Filipino scientists from the Ateneo de Manila University.

Instead of immersing entire sheets of metal into acidic solutions, the researchers applied microdroplets of acidic solution onto small aluminum surfaces and applied an . Just two volts of electricity—barely more than what’s found in a single AA household flashlight battery—was all that was needed to transform the metal into glass-like TAlOx.

The highly pathogenic avian influenza H5N1 is an emerging and unexpected threat to many wild animal species, which has implications for ecological processes, ecosystem services and conservation of threatened species. International collaboration and information-sharing is essential for surveillance, early diagnosis and the provision of financial and technical instruments to enable worldwide actions.

How can machine learning help determine the best times and ways to use solar energy? This is what a recent study published in Advances in Atmospheric Sciences hopes to address as a team of researchers from the Karlsruhe Institute of Technology investigated how machine learning algorithms can be used to predict and forecast weather patterns to enable more cost-effective approaches for using solar energy. This study has the potential to help enhance renewable energy technologies by fixing errors that are often found in current weather prediction models, leading to more efficient use of solar power by predicting when weather patterns will enable the availability of the Sun for solar energy needs.

For the study, the researchers used a combination of statistical methods and machine learning algorithms to help predict the most efficient times of day that photovoltaic (PV) power generation will achieve maximum production output. Their methods used what’s known as post-processing, which involves correcting weather forecasting errors before that data enters PV models, resulting in changing PV model predictions, resulting in establishing more accurate weather forecasting from machine learning algorithms.

“One of our biggest takeaways was just how important the time of day is,” said Dr. Sebastian Lerch, who is a professor at the Karlsruhe Institute of Technology and a co-author on the study. “We saw major improvements when we trained separate models for each hour of the day or fed time directly into the algorithms.”

What can a moon’s tidal friction teach us about its formation and evolution? This is what a recent study published in Science Advances hopes to address as a team of researchers at the University of California Santa Cruz investigated a connection between the spin rate and tidal energy on Saturn’s moon, Titan, to determine more about Titan’s interior. This study has the potential to help researchers better understand the internal processes of Titan, leading to better constraints on the existence of a subsurface ocean.

For the study, the researchers used a combination of data obtained by NASA’s now-retired Cassini spacecraft and a series of mathematical calculations to determine Titan’s tidal dissipation, which is the amount of tidal energy lost in an object from friction and other processes, and for which the only moons in the solar system this has been successfully been accomplished being the Earth’s Moon and Jupiter’s volcanic moon, Io. Better understanding a moon’s tidal dissipation helps researchers better understand its formation and evolution, which the researchers successfully estimated for Titan.

“Tidal dissipation in satellites affects their orbital and rotational evolution and their ability to maintain subsurface oceans,” said Dr. Brynna Downey, who is a postdoctoral researcher at the Southwest Research Institute in Colorado and lead author of the study. “Now that we have an estimate for the strength of tides on Titan, what does it tell us about how quickly the orbit is changing? What we discovered is that it’s changing very quickly on a geologic timescale.”