Scientists reveal what triggered Santorini'earthquake swarm' The swarm of tens of thousands of earthquakes near the Greek island of Santorini earlier this year was triggered by molten rock pumping through an underground channel over three months, scientists have discovered. They used physics and artificial intelligence to work out exactly what caused the more than 25,000 earthquakes, which travelled about 20km (12 miles) horizontally through the Earth's crust. They used each of the tremors as virtual sensors, then used artificial intelligence to analyse patterns associated with them. One of the lead researchers, Dr Stephen Hicks from UCL, said combining physics and machine learning in this way could help forecast volcanic eruptions. The seismic activity started to stir beneath the Greek islands of Santorini, Amorgos, and Anafi in January 2025.

Breakthroughs, discoveries, and DIY tips sent every weekday. Don't let the appetizing description fool you. When planetary scientists say the interior of Mars resembles a rocky road brownie more than a piece of buttery shortbread, the tasty metaphor masks billions of years of geological violence. In a re-examination of previous observations collected by NASA's decommissioned InSight probe, researchers have discovered that the Martian mantle is embedded with ancient fragments measuring as much as 2.5 miles wide. The data is detailed in a study published on August 28 in Nature.

--Dermatological care via telemedicine often lacks the rich context of in-person visits. Clinicians must make diagnoses based on a handful of images and brief descriptions, without the benefit of physical exams, second opinions, or reference materials. While many medical AI systems attempt to bridge these gaps with domain-specific fine-tuning, this work hypothesized that mimicking clinical reasoning processes could offer a more effective path forward. This study tested seven vision-language models on medical visual question answering across six configurations: baseline models, fine-tuned variants, and both augmented with either reasoning layers that combine multiple model perspectives, analogous to peer consultation, or retrieval-augmented generation that incorporates medical literature at inference time, serving a role similar to reference-checking. While fine-tuning degraded performance in four of seven models with an average 30% decrease, baseline models collapsed on test data. Clinical-inspired architectures, meanwhile, achieved up to 70% accuracy, maintaining performance on unseen data while generating explainable, literature-grounded outputs critical for clinical adoption. These findings demonstrate that medical AI succeeds by reconstructing the collaborative and evidence-based practices fundamental to clinical diagnosis. Fine-tuning large models on medical data, the standard approach to medical AI, assumes domain exposure produces clinical competence [1]. Y et dermatology models show 15% performance drops in real-world settings [2], and catastrophic forgetting causes models to generate outputs exclusively from their training data [3]. This brittleness suggests a fundamental mismatch between current approaches and clinical reasoning. Additionally, physician groups achieve 85.6% diagnostic accuracy versus 62.5% for individuals [4], as collaboration reduces cognitive load and bias [5]. However, logistical constraints force physicians to work alone, a problem telemedicine intensifies by eliminating physical exams, peer consultation, and immediate reference access [6].

Large Language Models (LLMs) have demonstrated superior results across a wide range of tasks, while retrieval has long been established as an effective means of obtaining task-relevant information for humans. Retrieval-augmented Generation (RAG) are known for their effectiveness in knowledge-intensive tasks by locating relevant information and placing it within the context window of the LLM. However, the relationship between retrievers and LLMs is still under-investigated. Most existing work treats the retriever and the LLM as independent components and leaves a gap between retrieving human-friendly information and assembling a LLM-friendly context. In this work, we examine a novel bridge model, validate the ranking and selection assumptions in retrievers in the context of RAG, and propose a training framework that chains together supervised and reinforcement learning to learn a bridge model. Empirical results demonstrate the effectiveness of our method in both question-answering and personalized generation tasks.

Scientists have discovered a hidden layer of Earth, which sits 100 miles below the surface and covers at least 44 percent of the planet. This previously unknown region of molten rock is part of the asthenosphere, located under tectonic plates in the upper mantle, which forms a soft boundary that allows the solid rock slabs to move. While the discovery is significant, it shatters long-held theories that molten rocks influence the asthenosphere's viscosity. Junlin Hua, with the University of Texas, Austin, said in a statement: 'When we think about something melting, we intuitively think that the melt must play a big role in the material's viscosity. 'But what we found is that even where the melt fraction is quite high, its effect on mantle flow is very minor.'

NASA's InSight lander has measured one of the biggest and longest marsquakes yet, which featured tremors of 4.2 magnitude lasting nearly an hour and a half, the space agency said. The robotic seismometre celebrated 1,000 days on the Red Planet on September 18, when it detected the largest tremor since it arrived at the Elysium Planitia in 2018. The 4.2 magnitude quake equals the largest detected so far on Mars, but on Earth that would be considered'light', with more than 10,000 earthquakes of that level detected every year, feeling like a light rumble that would make dishes shake. The lander was only able to make the measurement after efforts to clear dust from its solar panels earlier in the year - keeping the seismometre operating. The team took a counterintuitive approach to achieving this by sprinkling one solar panel with larger sand grains in the hope wind would blow it across the other panel and result in clearing enough of the dust to allow power to enter the device.

Scientists have finally been able to understand the crust underneath the surface of Mars. The research represents the first time that humanity has been able to start mapping the interior of another planet beyond our own Earth. The new research relied on data taken from Nasa's InSight mission, which has been looking for Marsquakes that reverberate across its surface. Using information about those quakes, researchers are able to understand what might be lurking beneath the Martian surface. Beneath the InSight landing site, the crust is either approximately 20 kilometres or 39 kilometres thick, according to an international research team led by geophysicist Dr Brigitte Knapmeyer-Endrun at the University of Cologne's Institute of Geology and Mineralogy and Dr Mark Panning at Jet Propulsion Laboratory, California Institute of Technology (Caltech).

Data beamed back to Earth from NASA's InSight lander suggests Mars' crust is composed of three cake-like layers. Anchored near Mars' equator, the robotic lander's super-sensitive seismometer, known as SEIS, has recorded hundreds of'marsquakes' in the past two years. Each quake emits two sets of seismic waves and analyzing the differences in how those waves move has allowed researchers to begin calculating the size and composition of the planet's crust, mantle and core. 'We have enough data to start answering some of these big questions,' Jet Propulsion Laboratory scientist Bruce Banerdt told Nature. Launched in 2018, the InSight mission marks the first time scientists have peered inside a planet other than Earth.

Circulating groundwater triggered a four-year-long swarm of tiny earthquakes that rumbled beneath the Southern California town of Cahuilla, researchers report in the June 19 Science. By training computers to recognize such faint rumbles, the scientists were able not only to identify the probable culprit behind the quakes, but also to track how such mysterious swarms can spread through complex fault networks in space and time. Seismic signals are constantly being recorded in tectonically active Southern California, says seismologist Zachary Ross of Caltech. Using that rich database, Ross and colleagues have been training computers to distinguish the telltale ground movements of minute earthquakes from other things that gently shake the ground, such as construction reverberations or distant rumbles of the ocean (SN: 4/18/19). The millions of tiny quakes revealed by this machine learning technique, he says, can be used to create high-resolution, 3-D images of what lies beneath the ground's surface in a particular region.

It sounds like a subway train rushing by. But it's something much more exotic: in all likelihood, the first "marsquake" ever recorded by humans. NASA's InSight mission detected the quake on April 6, four months after the lander's highly sensitive seismometer was installed on the Martian surface. The instrument had previously registered the howling winds of the red planet and the motions of the lander's robotic arm. But the shaking picked up this month is believed to be the first quake from Mars' interior.