Artificial Intelligence (AI), particularly large language models like GPT-4, has shown impressive performance on reasoning tasks. But does AI truly understand abstract concepts, or is it just mimicking patterns? A new study from the University of Amsterdam and the Santa Fe Institute reveals that while GPT models perform well on some analogy tasks, they fall short when the problems are altered, highlighting key weaknesses in AI’s reasoning capabilities. The work is published in Transactions on Machine Learning Research.

Analogical reasoning is the ability to draw a comparison between two different things based on their similarities in certain aspects. It is one of the most common methods by which human beings try to understand the world and make decisions. An example of analogical reasoning: cup is to coffee as soup is to??? (the answer being: bowl)

Large language models (LLMs) like GPT-4 perform well on various tests, including those requiring analogical reasoning. But can AI models truly engage in general, robust reasoning or do they over-rely on patterns from their training data? This study by language and AI experts Martha Lewis (Institute for Logic, Language and Computation at the University of Amsterdam) and Melanie Mitchell (Santa Fe Institute) examined whether GPT models are as flexible and robust as humans in making analogies.

A research team has identified different subtypes of white matter (WM) astrocytes, including a unique type with the ability to multiply and potentially aid in brain repair. Using single-cell RNA sequencing and spatial transcriptomics, the scientists mapped astrocyte diversity across different brain regions and species, providing the first detailed molecular profile of WM astrocytes.

The team was led by Dr. Judith Fischer-Sternjak from Helmholtz Munich and Ludwig-Maximilians-Universität (LMU) München, alongside Prof. Magdalena Götz from Helmholtz Munich, LMU and the Munich Cluster for Systems Neurology (SyNergy). The research is published in the journal Nature Neuroscience.

Unveiling white matter astrocyte diversity Astrocytes, known for their crucial role in supporting neurons and maintaining brain health, have been predominantly studied in gray matter (GM), which is involved in information processing. However, white matter astrocytes, which support long-range neural connections, remain poorly understood. This study fills a major knowledge gap by showing that WM astrocytes are not a uniform population but consist of distinct subtypes with specialized roles.

Is NASA still Moonbound, or will the next giant leap mean skipping straight to Mars?

Speculation is mounting that the Trump administration may scale back or cancel NASA’s Artemis missions following the departure of a key official and Boeing’s plans to lay off hundreds of employees working on its lunar rocket.

Late Wednesday, NASA abruptly announced the retirement of longtime associate administrator Jim Free, effective Saturday.

There are plenty of types of stars out there, but one stands out for being just a little weirder than the others. You might even say it’s strange. According to a paper from researchers at Guangxi University in China, the birth of one might have recently been observed for the very first time.

A strange star is a (so far theoretical) compact star that is so dense it literally breaks down regular parts of atoms (like neutrons) into their constituent quarks. Moreover, even those quarks (the up and down that comprise a neutron) get compressed into an even rarer type of quark called a strange quark—hence the name strange star.

Technically, the “strange” matter that a strange star would be composed of is a combination of up, down, and strange quarks. But, at least in theory, this mix of sub-hadronic particles could even be more stable than a traditional neutron star, which is similar to a strange star but doesn’t have enough gravity to break down the neutrons.

Lobster-eye satellite Einstein Probe captured the X-ray flash from a very elusive celestial pair. The discovery opens a new way to explore how massive stars interact and evolve, confirming the unique power of the mission to uncover fleeting X-ray sources in the sky.

The odd celestial couple consists of a big, hot star, more than 10 times larger than our Sun, and a small compact white dwarf, with a mass similar to our star. Only a handful of these systems have been found so far. And this the first time scientists could track the X-ray light coming from such a curious pair from its initial sudden flare-up to its fading away.

On 27 May 2024, the Wide-field X-ray Telescope (WXT) on Einstein Probe spotted X-rays coming from within our neighbour galaxy, the Small Magellanic Cloud (SMC). To uncover the origin of this new celestial beacon, labelled EP J0052, scientists pointed Einstein Probes’s Follow-up X-ray Telescope in that direction.

Back in 1971, a couple of British astronomers predicted the existence of a black hole at the center of our galaxy. And in 1974, other astronomers found it, naming it Sagittarius A*.

Since then, astronomers have discovered that a similar “supermassive black hole” sits at the center of almost every other large galaxy. In 2019, they took the first image of a supermassive black hole. Today, these exotic objects are a fundamental part of our understanding of how galaxies form and evolve.

But what of smaller astronomical bodies, like the Large Magellanic Cloud, a dwarf satellite galaxy that is expected to collide with the Milky Way in 2.4 billion years? Nobody is quite sure whether clouds like this might also house supermassive black holes.

The speed of light is often regarded as the ultimate cosmic speed limit, but researchers have now managed to slow it down dramatically—to just 61 kilometers per hour. This was achieved by using a Bose-Einstein condensate (BEC), a peculiar quantum state of matter that allows light to be slowed or even stopped entirely. This discovery, which builds on decades of research, has implications for quantum physics, computing, and information storage.

The Quantum Jelly Effect In everyday conditions, light moves at 299,792,458 meters per second in a vacuum, and its speed decreases slightly when passing through materials like glass or water. However, these reductions are relatively small. In contrast, when light travels through a Bose-Einstein condensate, it can be slowed to a near standstill.

A Bose-Einstein condensate is an exotic state of matter, first predicted by Albert Einstein and Satyendra Nath Bose, that occurs when a gas is cooled to temperatures just above absolute zero. Under these conditions, the atoms behave as a single quantum entity, exhibiting superfluidity and interacting with light in ways not seen in ordinary materials.

Newly discovered brain cells count each bite before sending the order to cease eating a meal. Columbia scientists have found specialized neurons in the brains of mice that order the animals to stop eating.

Though many feeding circuits in the brain are known to play a role in monitoring food intake, the neurons in those circuits do not make the final decision to cease eating a meal.

The neurons identified by the Columbia scientists, a new element of these circuits, are located in the brainstem, the oldest part of the vertebrate brain. Their discovery could lead to new treatments for obesity.

Researchers have developed a reactor that pulls carbon dioxide directly from the air and converts it into sustainable fuel, using sunlight as the power source.

The researchers, from the University of Cambridge, say their solar-powered reactor could be used to make fuel to power cars and planes, or the many chemical and pharmaceutical products we rely on. It could also be used to generate fuel in remote or off-grid locations.

Unlike most carbon capture technologies, the reactor developed by the Cambridge researchers does not require fossil-fuel-based power, or the transport and storage of carbon dioxide, but instead converts atmospheric CO2 into something useful using sunlight. The results are reported in the journal Nature Energy.

Asteroids that orbit close to the Earth inevitably cause us some anxiety due to the even remote possibility of a collision. But their proximity also offers ample opportunities to learn more about the universe. Ryugu, a 900-meter diameter asteroid in the Apollo belt, has recently proven useful in our search for signs of life’s precursors elsewhere in our solar system.

A team of researchers at Kyoto University have found evidence of salt minerals in samples recovered from Ryugu during the initial phase of Japan’s Hayabusa2 mission. The discovery of these deposits, containing sodium carbonate, halite, and sodium sulfates, suggest that liquid saline water once existed within a parent body of Ryugu.

Before examining the samples, the team expected that sample grains returned from the asteroid might contain substances not generally found in meteorites. They anticipated that these could be highly water-soluble materials, which readily react with moisture in Earth’s atmosphere and are difficult to detect unless examined in their pristine state as preserved in the vacuum of space.