<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>STEM Synapse</title><description>New papers, preprints and science news, explained simply, with every source one click away, alongside the pictures science makes.</description><link>https://stemsynapse.org/</link><language>en-gb</language><item><title>AI &amp; Machine Learning: Fast Catch-Up, Late Switching: Optimal Batch Size Scheduling via Functional Scaling Laws</title><link>https://stemsynapse.org/papers/2026-11-18/#sb-ai-ml</link><guid isPermaLink="false">2026-11-18/ai-ml</guid><description>Language models are trained on data fed in batches, and how large those batches should be at each stage is usually set by rule of thumb. The researchers used a mathematical description of how training error falls over time to find the best batch-size schedule when the amount of data is fixed. For easy tasks the batch should keep growing; for hard ones it should stay small for most of training and jump to large only near the end. After that late switch the model quickly catches up, because the noise built up from small batches fades fast. Tests on models of up to 1.1 billion parameters showed late switching beat constant and early-switch schedules while using less data.</description><pubDate>Wed, 18 Nov 2026 07:08:25 GMT</pubDate><category>AI &amp; Machine Learning</category></item><item><title>Tech &amp; Engineering: Adversarial Reinforcement Learning for Adaptive Eavesdropping in BB84 Quantum Key Distribution</title><link>https://stemsynapse.org/papers/2026-11-18/#sb-tech-engineering</link><guid isPermaLink="false">2026-11-18/tech-engineering</guid><description>Quantum key distribution lets two parties share a secret code with a guarantee from physics: anyone listening in disturbs the signal in a way that shows up as extra errors. Security checks usually assume the eavesdropper intercepts at a fixed rate. Here the eavesdropper was instead a simple learning program that watched the error rate and decided, one quantum bit at a time, whether to intercept. With 1% channel noise, it cut its chance of being detected from 99.4% under the best fixed strategy to about 0.28%, while still collecting around 10 correct key bits per round. The agents also learned to burst their attacks at the very end, a trick that randomised check timing removed.</description><pubDate>Wed, 18 Nov 2026 07:08:25 GMT</pubDate><category>Tech &amp; Engineering</category></item><item><title>Space, Astronomy &amp; Aerospace: Bayesian Inference of Stellar r-process Abundances</title><link>https://stemsynapse.org/papers/2026-11-18/#sb-space</link><guid isPermaLink="false">2026-11-18/space</guid><description>Many elements heavier than iron, including gold and uranium, are made in the rapid neutron-capture process, a burst of nuclear reactions whose exact settings in space are still debated. Old stars rich in these elements keep a record of it. The researchers built a statistical method that explains a star&apos;s pattern of elements as a mix of simulated production events, each with its own conditions. For one well-studied star, exactly two components were needed, one making the heavier elements and one the lighter ones, and adding more did not help. Two other old stars and the Sun&apos;s own heavy elements fit almost the same two components in different proportions, although the Sun&apos;s lightest heavy elements need an extra source.</description><pubDate>Wed, 18 Nov 2026 07:08:25 GMT</pubDate><category>Space, Astronomy &amp; Aerospace</category></item><item><title>Physics: Spin-Resolved Microscopy of ⁸⁷Sr SU(N) Fermi-Hubbard Systems</title><link>https://stemsynapse.org/papers/2026-11-18/#sb-physics</link><guid isPermaLink="false">2026-11-18/physics</guid><description>Physicists study magnetism by trapping ultracold atoms in a grid made of laser light and photographing them one by one, but this has mostly used atoms with two spin states, the quantum property behind magnetism. Strontium-87 atoms have ten, which allows richer kinds of magnetism that have never been watched atom by atom. The researchers built a microscope for strontium-87 that sees single atoms and, using laser pulses that act on each spin state separately, records which of the ten states every atom is in from a single snapshot. They checked it by watching single atoms&apos; spins turn in a magnetic field, and simulations show it should reveal how the spins of neighbouring atoms are linked.</description><pubDate>Wed, 18 Nov 2026 07:08:25 GMT</pubDate><category>Physics</category></item><item><title>Chemistry &amp; Materials Science: Sulfurization Breaking the Ferroelectric Polarization Bottleneck in PbTiO3</title><link>https://stemsynapse.org/papers/2026-11-18/#sb-chemistry-materials</link><guid isPermaLink="false">2026-11-18/chemistry-materials</guid><description>Ferroelectric materials hold an electric polarisation that can be flipped and kept without power, which makes them useful for memory chips. Lead titanate should be one of the strongest, but electric current leaking through it has kept that strength from showing in practice. The researchers made thin films of it with a cheap solution-based method and then treated them with sulfur. Untreated films already performed well, and sulfur treatment raised the polarisation they kept to about 1.8 times its original value, while the material stayed ferroelectric at higher temperatures. Calculations suggest sulfur stretches the crystal in a way that favours the polar arrangement. The authors see the method as promising for ferroelectric devices.</description><pubDate>Wed, 18 Nov 2026 07:08:25 GMT</pubDate><category>Chemistry &amp; Materials Science</category></item><item><title>Biology &amp; Ecology: Groundfish With Diverse Life Histories Increase in Size and Abundance With Proximity to Spatial Protections</title><link>https://stemsynapse.org/papers/2026-11-18/#sb-biology-ecology</link><guid isPermaLink="false">2026-11-18/biology-ecology</guid><description>Closing patches of sea to fishing should let fish grow bigger and more numerous inside them and spill over into nearby waters, but that is hard to show in real data. The researchers used 18 years of surveys of 28 kinds of bottom-dwelling fish, collected by four Indigenous Nations on the Pacific coast of Canada, inside and up to 160 kilometres outside seven Rockfish Conservation Areas. Fish were largest and most abundant inside the protected areas and became bigger and more common closer to them, and these effects strengthened as the areas aged. Fast-breeding species increased in number sooner, as expected, but slow-breeding species grew larger sooner, the opposite of the prediction.</description><pubDate>Wed, 18 Nov 2026 07:08:25 GMT</pubDate><category>Biology &amp; Ecology</category></item><item><title>Health &amp; Biomedical: Semaglutide for the Treatment of Obesity in Children 6 to &lt;12 Years of Age</title><link>https://stemsynapse.org/papers/2026-11-18/#sb-health-biomedical</link><guid isPermaLink="false">2026-11-18/health-biomedical</guid><description>Semaglutide, a weekly injected drug, is approved for obesity in teenagers and adults, but there has been little evidence on whether it helps younger children. In this trial, 167 children aged 6 to 11 with obesity were randomly given weekly semaglutide or a dummy injection for two years, and all of them received advice on lifestyle changes. After 68 weeks, body-mass index had fallen by 13.5% on average with semaglutide and risen by 2.7% with the dummy injection. The difference was still large at two years. Side effects were reported in nearly all children in both groups, those on semaglutide and those on the dummy injection.</description><pubDate>Wed, 18 Nov 2026 07:08:25 GMT</pubDate><category>Health &amp; Biomedical</category></item><item><title>Climate, Energy &amp; Environment: Distinct Methane-Cycling Pathways Drive Variable Surface Flux in Freshwater Wetlands</title><link>https://stemsynapse.org/papers/2026-11-18/#sb-climate-energy</link><guid isPermaLink="false">2026-11-18/climate-energy</guid><description>Freshwater wetlands are the largest natural source of methane, a potent greenhouse gas, making up about a third of global emissions, yet what controls how much each wetland releases is unclear. Microbes both make methane and consume it. The researchers studied six different wetlands in the Prairie Pothole Region of North America, measuring which microbial genes were active, the chemistry of water in the soil, methane underground and the gas escaping at the surface. Wetlands where microbes were busier using nitrate and sulfate made less methane. Yet how much methane microbes produced did not predict how much reached the air; that depended on local conditions shaping the microbes that eat it.</description><pubDate>Wed, 18 Nov 2026 07:08:25 GMT</pubDate><category>Climate, Energy &amp; Environment</category></item><item><title>Psychology &amp; Cognitive Science: From media to dreams: The effects of the Russo-Ukrainian war on threatening dream content in a Finnish dream diary sample</title><link>https://stemsynapse.org/papers/2026-11-18/#sb-psychology-cogsci</link><guid isPermaLink="false">2026-11-18/psychology-cogsci</guid><description>One theory holds that dreams evolved to rehearse threats, especially those our ancestors faced, such as physical violence, more than threats like disease. If so, war should change dreams more than a pandemic does. The researchers compared two Finnish dream-diary studies run the same way: 567 dreams from 45 people recorded after Russia&apos;s full-scale invasion of Ukraine, and 1,128 dreams from 85 people recorded under the strictest COVID-19 restrictions. Dreams after the invasion contained more threats, above all of physical aggression, even though participants knew the war only through the media. Dreams that mentioned war directly held the most such threats. The threats were not more severe, though, contrary to what the researchers expected.</description><pubDate>Wed, 18 Nov 2026 07:08:25 GMT</pubDate><category>Psychology &amp; Cognitive Science</category></item><item><title>Math &amp; Computer Science Foundations: Minimum cores in lattice graphs: zonotope structure and quartic growth</title><link>https://stemsynapse.org/papers/2026-11-18/#sb-math-cs-foundations</link><guid isPermaLink="false">2026-11-18/math-cs-foundations</guid><description>Picture an endless grid where each point is linked to every point within a set distance. A core is a group of points in which every member has at least as many links inside the group as outside it, and the question is how small such a group can be. The author shows that every core&apos;s outline must include a particular many-sided shape fixed by the grid, and that on a flat grid the smallest core grows with the fourth power of the linking distance. When each point is linked only to its eight surrounding points, the smallest core has exactly 12 points. Finding cores in general networks is also shown to be computationally hard, settling an earlier conjecture.</description><pubDate>Wed, 18 Nov 2026 07:08:25 GMT</pubDate><category>Math &amp; Computer Science Foundations</category></item><item><title>AI &amp; Machine Learning: PhysDox: Benchmarking LLMs on Physical Feasibility Auditing of Physiological Sensing Protocols</title><link>https://stemsynapse.org/papers/2026-11-17/#sb-ai-ml</link><guid isPermaLink="false">2026-11-17/ai-ml</guid><description>Language models are increasingly asked to help design experiments, and they write plans that read well, yet a fluent plan can still be physically impossible to carry out. The researchers built a test of 683 expert-checked plans for measuring signals from the human body, each graded as sound, slightly flawed or fatally flawed, and asked six models to judge them. The best reached a score of only 53 out of 100 at spotting the problems. The models tended to mistake a plan that lists every step for one that would work, and they missed unstated physical limits about twice as often as plain hardware errors. The authors conclude that checking feasibility takes real physical reasoning rather than recall.</description><pubDate>Tue, 17 Nov 2026 07:07:44 GMT</pubDate><category>AI &amp; Machine Learning</category></item><item><title>Tech &amp; Engineering: Design and Experimental Evaluation of a Low-Cost Immersion Cooling System</title><link>https://stemsynapse.org/papers/2026-11-17/#sb-tech-engineering</link><guid isPermaLink="false">2026-11-17/tech-engineering</guid><description>Sinking computer hardware in oil is one way to cool crowded data centres, but rigs for studying it are usually costly. The researchers built an open design from parts available in India for about 380 US dollars: a heater standing in for a server sits in a glass tank of mineral oil, a pump pushes the oil through a fan-cooled radiator, and a small microcontroller logs the temperatures. At full load the oil stayed only 8 degrees Celsius above room temperature while the heater surface ran far hotter. Almost all the resistance to heat flow sat where the heater meets the slowly moving oil, so the cooling loop itself has capacity to spare. The plans, code and data are public.</description><pubDate>Tue, 17 Nov 2026 07:07:44 GMT</pubDate><category>Tech &amp; Engineering</category></item><item><title>Space, Astronomy &amp; Aerospace: Magnetic Reconnection Governs the Entire Lifecycle of Solar Eruptive Magnetic Flux Rope</title><link>https://stemsynapse.org/papers/2026-11-17/#sb-space</link><guid isPermaLink="false">2026-11-17/space</guid><description>Before many solar eruptions, a twisted bundle of magnetic field called a flux rope sits in the Sun&apos;s atmosphere, and what makes it suddenly blast outward is still argued over. The usual explanation is an instability that sets in once the rope rises past a certain height, with magnetic reconnection, the breaking and rejoining of field lines, in a supporting role. Using detailed three-dimensional computer simulations, the researchers grew a flux rope and watched it rise past that height without erupting. It erupted only once the current layer beneath it became thin enough for reconnection to begin, and reconnection then supplied most of the push. The authors conclude that, in this model, reconnection builds, triggers and drives the eruption.</description><pubDate>Tue, 17 Nov 2026 07:07:44 GMT</pubDate><category>Space, Astronomy &amp; Aerospace</category></item><item><title>Physics: Composition-dependent ultralight dark matter detection with a tetrahedral superconducting gravity gradiometer</title><link>https://stemsynapse.org/papers/2026-11-17/#sb-physics</link><guid isPermaLink="false">2026-11-17/physics</guid><description>Some theories propose that dark matter is an extremely light field that ripples through space and tugs on different materials with slightly different strength. Catching such a tug needs very sensitive instruments, and the researchers calculate that suitable ones are already being designed: superconducting gravity gradiometers, which track tiny differences in how four test masses accelerate and are planned as gravitational-wave detectors on the ground and on the Moon. Making two of the four masses from copper and two from lead would add a dark matter signal to the same readout, with no extra sensor. That signal points along a direction fixed in advance by where the metals sit, which a gravitational wave cannot mimic.</description><pubDate>Tue, 17 Nov 2026 07:07:44 GMT</pubDate><category>Physics</category></item><item><title>Chemistry &amp; Materials Science: Self‐Sustaining Lattice‐Oxygen Cycling Enables Durable Industrial‐Level Seawater Electrolysis</title><link>https://stemsynapse.org/papers/2026-11-17/#sb-chemistry-materials</link><guid isPermaLink="false">2026-11-17/chemistry-materials</guid><description>Splitting seawater with electricity could make hydrogen fuel without drawing on fresh water, but fast catalysts tend to fall apart, worn down by chloride from the salt and by gaps that open in their own structure as they give up oxygen. The researchers made a disordered catalyst of ruthenium, iron and cobalt dotted with silver. The silver forms a surface layer that keeps chloride away and helps refill the missing oxygen. In natural seawater it ran at an industrial current for a combined total of more than 9,500 hours, and a complete electrolyser built with it also ran without a break. The authors present the approach as a general route to durable hydrogen production from seawater.</description><pubDate>Tue, 17 Nov 2026 07:07:44 GMT</pubDate><category>Chemistry &amp; Materials Science</category></item><item><title>Biology &amp; Ecology: Conserved Upper Thermal Limits and Small Safety Margins in Soil Copiotrophic Bacteria</title><link>https://stemsynapse.org/papers/2026-11-17/#sb-biology-ecology</link><guid isPermaLink="false">2026-11-17/biology-ecology</guid><description>How microbes will cope with rising temperatures is one of the open questions in climate models, and soil bacteria are a large part of it. The researchers grew 251 isolates of fast-growing soil bacteria from 18 genera, collected from topsoils across California, at a range of temperatures. Their best growing temperature and their upper heat limit were set mainly by evolutionary lineage, with little link to the local climate, and variation in these limits narrowed above 40 degrees Celsius. Bacteria from hot places with swinging temperatures had little margin to spare. The authors suggest many of these bacteria may already be poorly matched to local heat, and that their ability to adapt to warming may be limited.</description><pubDate>Tue, 17 Nov 2026 07:07:44 GMT</pubDate><category>Biology &amp; Ecology</category></item><item><title>Climate, Energy &amp; Environment: Seasonal Asynchrony of Microbial Biomass and Respiration Across Terrestrial Ecosystems</title><link>https://stemsynapse.org/papers/2026-11-17/#sb-climate-energy</link><guid isPermaLink="false">2026-11-17/climate-energy</guid><description>Soil microbes hold carbon in their own cells and release it as they breathe, and climate models usually assume that more microbes means more breathing out. Whether that holds through the seasons was unclear. The researchers pooled seasonal measurements from 323 studies and added detailed observations at individual sites. The times of year with the most microbial biomass were often not the times with the highest respiration. Biomass rose and fell with competition from plants, while respiration followed temperature. The authors stress that the evidence is correlational, but say models of how soils release carbon should treat the two separately as seasonal swings in climate grow.</description><pubDate>Tue, 17 Nov 2026 07:07:44 GMT</pubDate><category>Climate, Energy &amp; Environment</category></item><item><title>Psychology &amp; Cognitive Science: Hunger and Concentration in Daily Life: Inter-Individual Differences and Eating-Related Variation but No Robust Average Association</title><link>https://stemsynapse.org/papers/2026-11-17/#sb-psychology-cogsci</link><guid isPermaLink="false">2026-11-17/psychology-cogsci</guid><description>Does being hungry make it harder to concentrate? Hunger is a frequent bodily signal that could pull attention away from a task, but people usually eat to quiet it, which makes its everyday effect hard to pin down. Here 64 people reported their concentration, hunger and recent eating six times a day for four weeks. Across the group there was no reliable link: on average, hungrier moments were not less focused ones. The link did vary a lot from person to person, and in exploratory analyses hunger went with poorer concentration after breakfast and lunch. The authors suggest that everyday focus depends less on hunger itself than on the person and the meal context.</description><pubDate>Tue, 17 Nov 2026 07:07:44 GMT</pubDate><category>Psychology &amp; Cognitive Science</category></item><item><title>Math &amp; Computer Science Foundations: Positive Dual Certificates for Real Separated Subset Sums: Chamber Optimality of the Conway-Guy and Bohman Sets, and Exact Values up to Eight</title><link>https://stemsynapse.org/papers/2026-11-17/#sb-math-cs-foundations</link><guid isPermaLink="false">2026-11-17/math-cs-foundations</guid><description>Pick some positive numbers so that all the sums you can make from them differ by at least 1: how small can the largest number be? This is a version, allowing any real numbers, of a famous Erdős puzzle about whole numbers, and a construction by Conway and Guy is a well-known candidate answer. The author splits the problem into regions where it becomes a straightforward optimization and proves that the Conway-Guy set is the unique best choice within its region for every size. A computer search with exact checks then shows that for up to eight numbers, real numbers do no better than whole numbers; with eight, the largest must be 84.</description><pubDate>Tue, 17 Nov 2026 07:07:44 GMT</pubDate><category>Math &amp; Computer Science Foundations</category></item><item><title>AI &amp; Machine Learning: Goal Encoding in LLMs</title><link>https://stemsynapse.org/papers/2026-11-16/#sb-ai-ml</link><guid isPermaLink="false">2026-11-16/ai-ml</guid><description>When a language model works through a task with several steps, does it hold a goal internally before it acts? The researchers gave models coding tasks made of several sub-tasks and looked inside. They found the next sub-goal recorded in the output of a small set of attention heads, the components a model uses to pull in information from earlier text, and could read it out in plain words. Moving that signal into a different setting, such as another programming language, triggered the same operation there. Because such a model cannot carry an internal state forward for long, the authors propose that it keeps re-reading the goal from its instructions and checking progress, an idea they test with targeted experiments.</description><pubDate>Mon, 16 Nov 2026 07:06:31 GMT</pubDate><category>AI &amp; Machine Learning</category></item><item><title>AI &amp; Machine Learning: Negative-Only Policy Optimization for One-Sided Verifiable Rewards</title><link>https://stemsynapse.org/papers/2026-11-16/#sb-ai-ml</link><guid isPermaLink="false">2026-11-16/ai-ml/arxiv:2611.17287</guid><description>One way to train AI models is to let them attempt problems and reward the answers an automatic checker confirms as right. In many real cases no such checker exists: cheap tests, like a handful of unit tests for code, can prove an answer wrong but cannot prove it right. Rewarding everything that passes then teaches the model shortcuts that slip through the tests. The authors propose learning only from answers known to be wrong: the model is steered away from confirmed failures, and answers that pass are treated as unknown. On coding, logic and maths benchmarks the method solved more problems than existing approaches, and the gain held when it was carried over to other benchmarks.</description><pubDate>Mon, 16 Nov 2026 07:06:31 GMT</pubDate><category>AI &amp; Machine Learning</category></item><item><title>Tech &amp; Engineering: Implementation of a 1D Gaussian-modulated CV-QKD under finite-size parameter estimation</title><link>https://stemsynapse.org/papers/2026-11-16/#sb-tech-engineering</link><guid isPermaLink="false">2026-11-16/tech-engineering</guid><description>Quantum key distribution lets two parties share a secret encryption key using light, with security that rests on physics rather than on hard mathematics. A leading version encodes the key in two properties of a light wave, which places heavy demands on the sending equipment. Encoding in only one property is simpler, but a complete working system had been hard to build. The researchers built one, running over 12 km of ordinary optical fibre with separate lasers at each end and the full chain of processing needed to distil a secure key. It produced 0.364 megabits of secret key per second under realistic assumptions about limited data. That figure shows what the simpler design costs in performance.</description><pubDate>Mon, 16 Nov 2026 07:06:31 GMT</pubDate><category>Tech &amp; Engineering</category></item><item><title>Space, Astronomy &amp; Aerospace: Effects of Lensing by Sgr A* on Semicoherent Searches for Continuous Gravitational Waves. I. Signal Model and Mismatch Studies</title><link>https://stemsynapse.org/papers/2026-11-16/#sb-space</link><guid isPermaLink="false">2026-11-16/space</guid><description>Spinning neutron stars are expected to give off faint, steady gravitational waves, and astronomers search for them toward the centre of the Milky Way. For a star behind the giant black hole there, gravity would bend the waves along two paths, and the two copies would overlap and interfere. As the star moves, the interference shifts in ways the search templates leave out. The authors added simulated signals, with and without this effect, to noise and ran a standard search on them. Magnification helped detection, fast sideways motion took some of the gain away, and faint signals were affected more than strong ones. Limits from such searches therefore cannot simply be rescaled; the full distorted signal has to be simulated.</description><pubDate>Mon, 16 Nov 2026 07:06:31 GMT</pubDate><category>Space, Astronomy &amp; Aerospace</category></item><item><title>Physics: Circular phonons generate negligible magnetization in insulators</title><link>https://stemsynapse.org/papers/2026-11-16/#sb-physics</link><guid isPermaLink="false">2026-11-16/physics</guid><description>Laser light can make the atoms in a crystal move in tiny circles, and some experiments have suggested that this motion turns a non-magnetic material into a strong magnet. The evidence was indirect: it came from how the crystal twisted light passing through it, which can be read in more than one way. Here the researchers also measured the magnetic field just outside a silicon carbide crystal while driving it with intense infrared light. The twisting of light was as large as in earlier reports, but the field outside showed that any magnetism was below 0.001 tesla, far too small to explain it. Whatever twists the light, it is not a large magnetization.</description><pubDate>Mon, 16 Nov 2026 07:06:31 GMT</pubDate><category>Physics</category></item><item><title>Chemistry &amp; Materials Science: Ultra‐Stable Phosphide Barrier Layer for Thermoelectric Bismuth Telluride</title><link>https://stemsynapse.org/papers/2026-11-16/#sb-chemistry-materials</link><guid isPermaLink="false">2026-11-16/chemistry-materials</guid><description>Bismuth telluride is a widely used material for devices that turn a temperature difference directly into electricity, but generators built from it have been held back by the lack of a reliable barrier layer, the thin coating that keeps the material from reacting where it is joined to other parts. The researchers propose a family of compounds called metal phosphides and pick out nickel phosphide. It reacts very little with bismuth telluride, nickel moves through it slowly, and it bonds strongly at the atomic scale. The joint stayed stable for 90 days at about 250 °C, the longest reported for any such barrier. A test generator built with it turned up to 6.9% of the heat into electricity.</description><pubDate>Mon, 16 Nov 2026 07:06:31 GMT</pubDate><category>Chemistry &amp; Materials Science</category></item><item><title>Biology &amp; Ecology: Parental effects can be transmitted through the rhizosphere microbiome</title><link>https://stemsynapse.org/papers/2026-11-16/#sb-biology-ecology</link><guid isPermaLink="false">2026-11-16/biology-ecology</guid><description>Plants can pass on the effects of what they lived through, such as crowding by neighbours, to their offspring. Known routes include chemical marks on DNA; whether the microbes around the roots are another route was not known. The researchers grew eight genetic lines of common reed with or without a competing species, then raised their offspring in identical conditions. Offspring of parents that had faced competition grew in a more cautious, resource-saving way. They also had different communities of bacteria and fungi around their roots, and statistical modelling suggested these microbes may explain part of the change in the offspring. Parental experience may therefore shape the next generation through soil life as well as directly.</description><pubDate>Mon, 16 Nov 2026 07:06:31 GMT</pubDate><category>Biology &amp; Ecology</category></item><item><title>Climate, Energy &amp; Environment: Temperature-Dependent Decay of Infectivity of Infectious Salmon Anemia Virus (ISAV) in Raw Seawater</title><link>https://stemsynapse.org/papers/2026-11-16/#sb-climate-energy</link><guid isPermaLink="false">2026-11-16/climate-energy</guid><description>Infectious salmon anemia virus, a distant relative of the flu virus, threatens salmon farms worldwide and also wild fish. Controlling it depends on knowing how long the virus stays dangerous once it is loose in the sea. The researchers kept the virus in natural seawater at five temperatures from 4 to 17 °C and, after set times, tested whether it could still infect young salmon. The warmer the water, the faster the virus lost its ability to infect. In a model of how the virus drifts, the average distance over which fish could be exposed fell by about 40% as the sea warmed from 4 to 17 °C. The findings bear on how salmon farms guard against the disease.</description><pubDate>Mon, 16 Nov 2026 07:06:31 GMT</pubDate><category>Climate, Energy &amp; Environment</category></item><item><title>Psychology &amp; Cognitive Science: Environmental Preference and Context-Supported Episodic Recall</title><link>https://stemsynapse.org/papers/2026-11-16/#sb-psychology-cogsci</link><guid isPermaLink="false">2026-11-16/psychology-cogsci</guid><description>Do people remember things better when they come across them in places they like? Little research had looked. Here, 60 volunteers explored four virtual environments, some urban and some natural, some regular in layout and some irregular. They were then tested on what they remembered, straight away and after a delay, and asked which places they liked. A general liking for nature or for orderly spaces made no difference to memory. But in tests that used the setting as a cue, people recalled more objects from the particular environments they preferred. The authors suggest that how a place affects the mind may depend on how each person feels about it, not only on its design.</description><pubDate>Mon, 16 Nov 2026 07:06:31 GMT</pubDate><category>Psychology &amp; Cognitive Science</category></item><item><title>Math &amp; Computer Science Foundations: Combinatorial Bernoulli Factories with Polynomially Many Coins</title><link>https://stemsynapse.org/papers/2026-11-16/#sb-math-cs-foundations</link><guid isPermaLink="false">2026-11-16/math-cs-foundations</guid><description>Suppose you have coins whose odds are unknown and want to use them to make a random choice among complicated options, such as ways of pairing people up, that reflects those odds exactly. A procedure that does this is called a Bernoulli factory. An earlier one could need an astronomical number of coin flips. The authors give one algorithm, covering every problem where such a procedure can exist, whose expected number of flips grows at a manageable rate with the size of the problem and with how lopsided the coins are. For pairings of n items it needs roughly n to the eighth power flips, where the earlier method needed at least n to the power 2n minus 1.</description><pubDate>Mon, 16 Nov 2026 07:06:31 GMT</pubDate><category>Math &amp; Computer Science Foundations</category></item><item><title>AI &amp; Machine Learning: Weisfeiler-Lehman Meets Optimal Transport: Theory, Algorithms, and Implications</title><link>https://stemsynapse.org/papers/2026-11-15/#sb-ai-ml</link><guid isPermaLink="false">2026-11-15/ai-ml</guid><description>How do you measure how different two networks are? A classic test, named after Weisfeiler and Lehman, answers only yes or no: it checks whether two networks can be told apart by repeatedly comparing each point&apos;s neighbours. This paper turns the test into a distance, a number saying how far apart two networks are, which is zero exactly when the test cannot tell them apart. The distance can be computed efficiently and picks up finer differences than a leading existing method. From it the authors design a type of neural network that can, in principle, learn any continuous function of networks. The distance also turns out to be linked to a branch of mathematics about moving probability distributions.</description><pubDate>Sun, 15 Nov 2026 07:08:12 GMT</pubDate><category>AI &amp; Machine Learning</category></item><item><title>Tech &amp; Engineering: Reconfiguring System-Level Protection Through Bounded Delegation in RISC-V MCUs</title><link>https://stemsynapse.org/papers/2026-11-15/#sb-tech-engineering</link><guid isPermaLink="false">2026-11-15/tech-engineering</guid><description>The small chips inside embedded devices wall off parts of their memory so that one piece of software cannot tamper with another. Increasingly those walls need to move while the device is running. Existing chips can limit who may change the rules, but not what rules they may then write, so a permitted program can overreach. The authors propose a scheme in which each piece of software owns a region and can only make or hand on rules inside it, with every change checked before it takes effect. Built into an open-source chip design, it did not slow ordinary memory access, and switching rules during an update was 89.5% faster than routing the change through a supervising software layer.</description><pubDate>Sun, 15 Nov 2026 07:08:12 GMT</pubDate><category>Tech &amp; Engineering</category></item><item><title>Space, Astronomy &amp; Aerospace: The full Gaia colour-magnitude diagram with white dwarfs reveals a declining Milky Way star formation history</title><link>https://stemsynapse.org/papers/2026-11-15/#sb-space</link><guid isPermaLink="false">2026-11-15/space</guid><description>White dwarfs are the dense, slowly cooling remnants of stars like the Sun, and because they cool predictably they record when stars formed. They are usually modelled separately from ordinary stars. The authors built one model of the Milky Way&apos;s stars that must explain both at once, and tuned it against nearly eight million stars within about 3,300 light-years measured by the Gaia telescope, 110,000 of them white dwarfs. It reproduces the count of each kind of star within 3%, where an earlier model tuned only on ordinary stars predicted 81% too many white dwarfs. In the history it recovers, star formation peaks early and declines steadily, with two later bursts about 2.4 billion and 0.5 billion years ago.</description><pubDate>Sun, 15 Nov 2026 07:08:12 GMT</pubDate><category>Space, Astronomy &amp; Aerospace</category></item><item><title>Physics: Bound States of Two D0-Branes and the Three-Graviton Coupling</title><link>https://stemsynapse.org/papers/2026-11-15/#sb-physics</link><guid isPermaLink="false">2026-11-15/physics</guid><description>M-theory is a proposed framework for uniting gravity with quantum physics, and one conjecture says it can be captured by something far simpler: the quantum mechanics of a set of matrices. Testing that conjecture means solving the matrix model, which is hard. The authors solved its simplest case numerically, finding the lowest-energy state, which the conjecture identifies with a graviton, the hypothetical particle of gravity. Far from its centre, the shape of that state encodes how strongly three gravitons interact, a number already calculated another way. The numerical solution reproduces that number with high accuracy. The authors also show why: a symmetry of the model, supersymmetry, protects the value from corrections, which lets them derive it directly.</description><pubDate>Sun, 15 Nov 2026 07:08:12 GMT</pubDate><category>Physics</category></item><item><title>Chemistry &amp; Materials Science: Pulsed Electrosynthesis of Conformal High‐Entropy Metallic Glass Skins for Durable Industrial Water Electrolysis</title><link>https://stemsynapse.org/papers/2026-11-15/#sb-chemistry-materials</link><guid isPermaLink="false">2026-11-15/chemistry-materials</guid><description>Splitting water with electricity makes hydrogen, but electrodes must survive harsh conditions at the high currents industry needs. Metallic glasses, metals whose atoms are jumbled rather than arranged in crystals, resist corrosion well, but mixtures of several metals tend to crystallise instead. The researchers made such a glass at room temperature, in water: very short, intense pulses of current deposit the metals too quickly for crystals to form, leaving a skin about 100 nanometres thick on the electrode. Electrodes coated this way produced hydrogen at high current for more than 1,300 hours, while crystalline versions failed within about 200. In a complete water-splitting device they ran for 1,200 hours and outperformed a standard made with precious metals.</description><pubDate>Sun, 15 Nov 2026 07:08:12 GMT</pubDate><category>Chemistry &amp; Materials Science</category></item><item><title>Biology &amp; Ecology: Graphical Approach of the Interplay of Eco‐Evolutionary Dynamics and Coexistence</title><link>https://stemsynapse.org/papers/2026-11-15/#sb-biology-ecology</link><guid isPermaLink="false">2026-11-15/biology-ecology</guid><description>Species evolve while they compete, and that evolution can change whether they are able to go on living side by side. The two processes have been hard to view together. The authors propose a way to draw both on one diagram. Coexistence is shown as angles: one for how much room the whole community has to coexist, and one per species for how well it resists being pushed out. An evolutionary change then appears as an arrow across the same picture, so its effect on coexistence can be read off. Using it, the authors show when evolution helps coexistence and when it harms it, and find that evolution often leaves even the evolving species more vulnerable.</description><pubDate>Sun, 15 Nov 2026 07:08:12 GMT</pubDate><category>Biology &amp; Ecology</category></item><item><title>Health &amp; Biomedical: Sex and generation shape the genetic architecture of smoking initiation</title><link>https://stemsynapse.org/papers/2026-11-15/#sb-health-biomedical</link><guid isPermaLink="false">2026-11-15/health-biomedical</guid><description>Whether a person starts smoking is influenced partly by their genes, and attitudes to smoking have changed enormously over the past century. Did the genetic influence change too? Using data from nearly seven million people of European, East Asian, African American and Latino ancestry, the authors compared the genetics of starting to smoke in women and men and in older and younger generations. Genes explained more of the differences between people among women than among men, and more in younger generations than in older ones; ancestry made no such difference. The patterns also tracked anti-smoking policies and changing social norms. The authors conclude that the social setting shapes how genes influence behaviour.</description><pubDate>Sun, 15 Nov 2026 07:08:12 GMT</pubDate><category>Health &amp; Biomedical</category></item><item><title>Climate, Energy &amp; Environment: Belowground Productivity in Grasslands and Savannas Depends on Shifting Fire and Herbivory Regimes</title><link>https://stemsynapse.org/papers/2026-11-15/#sb-climate-energy</link><guid isPermaLink="false">2026-11-15/climate-energy</guid><description>In grasslands and savannas most plant growth happens out of sight, in roots, yet little was known about how fire and grazing animals affect it in different climates. The researchers measured root growth at seven long-running experiments in Africa and North America, from dry to fairly wet, where plots had been burned often, burned sometimes or protected from fire, and at six sites either open to large grazing animals or fenced off. Root growth was greater where water was more plentiful. Fire and grazing increased it, most of all at the wetter sites, where they held back trees and shrubs and favoured grasses. Drier sites changed little. The authors say these effects matter for how much carbon such landscapes store.</description><pubDate>Sun, 15 Nov 2026 07:08:12 GMT</pubDate><category>Climate, Energy &amp; Environment</category></item><item><title>Psychology &amp; Cognitive Science: Fast Eyes, Sharp Eyes, or Good Eyes? What Links Visual Perception to Higher-Order Cognition</title><link>https://stemsynapse.org/papers/2026-11-15/#sb-psychology-cogsci</link><guid isPermaLink="false">2026-11-15/psychology-cogsci</guid><description>People who do well on simple visual tests also tend to do well on memory and reasoning tests. Why? One idea is that they take in visual information faster; another, that they see more clearly through visual clutter; a third, that they have a better visual system in general. The researchers gave 79 people versions of the same tests that limited either viewing time or clarity, plus four memory and reasoning tests, and used a statistical model to separate the three abilities. Neither speed nor clarity was related to the thinking tests. Only a general visual ability, shared across every visual test, was strongly related to them. The result supports the third idea, which the authors call good eyes.</description><pubDate>Sun, 15 Nov 2026 07:08:12 GMT</pubDate><category>Psychology &amp; Cognitive Science</category></item><item><title>Math &amp; Computer Science Foundations: Integer Programming with Polyhedral Constraints on the Image</title><link>https://stemsynapse.org/papers/2026-11-15/#sb-math-cs-foundations</link><guid isPermaLink="false">2026-11-15/math-cs-foundations</guid><description>Integer programming is the problem of finding the best whole-number solution under a set of constraints, and it underlies scheduling, packing and many other planning tasks. This paper extends it to cases where the combined resources a solution uses must themselves satisfy further constraints. The authors give an algorithm, using random choices, whose running time matches the best known for ordinary integer programming, and improves on an earlier method for a more general setting. It yields exact algorithms for versions of the knapsack problem with extra conditions on what is used. It also resolves an open problem about a classic question: whether copies of given numbers can be added up to reach a target.</description><pubDate>Sun, 15 Nov 2026 07:08:12 GMT</pubDate><category>Math &amp; Computer Science Foundations</category></item><item><title>AI &amp; Machine Learning: Multimodal cell foundation models require synergistic integration beyond alignment</title><link>https://stemsynapse.org/papers/2026-11-14/#sb-ai-ml</link><guid isPermaLink="false">2026-11-14/ai-ml</guid><description>Biologists have AI models trained on single kinds of cell data, such as microscope images of tissue or readouts of which genes are active. Datasets pairing both kinds are scarce, so a common shortcut joins two ready-made models. Does the pair reveal anything the stronger model did not know? The authors introduce a score for the information that appears only when the two kinds of data are considered together. They show that the usual way of joining models mostly picks up what the two already share and misses what only the pair reveals. Across ten joining methods, a single model was enough for simple tasks, while defining complex cell neighbourhoods gained from methods built for the extra information.</description><pubDate>Sat, 14 Nov 2026 07:07:07 GMT</pubDate><category>AI &amp; Machine Learning</category></item><item><title>Tech &amp; Engineering: U.S. Data Center Growth Through 2030: A Critical Review of Grid Adequacy and Transmission Expansion Planning</title><link>https://stemsynapse.org/papers/2026-11-14/#sb-tech-engineering</link><guid isPermaLink="false">2026-11-14/tech-engineering</guid><description>Can the US electricity system keep up with the data centers planned through 2030? This review of public forecasts and studies argues that the headline numbers cannot answer the question. Forecasts mix three different things: average demand over a year, peak demand, and the size of connection a site requests. And a national total of generating capacity does not show a dependable surplus, because what can be delivered depends on plant availability, regional supply, transmission bottlenecks and project timing. Data centers also behave unusually: a fault on the grid can make many switch to backup power at once and return later. The review finds that this behaviour is rarely built into the studies used to plan new power lines.</description><pubDate>Sat, 14 Nov 2026 07:07:07 GMT</pubDate><category>Tech &amp; Engineering</category></item><item><title>Physics: Energy exchange confined to the perturbations: the frozen dark sector confronted with Planck, DESI, supernovae and cosmic shear</title><link>https://stemsynapse.org/papers/2026-11-14/#sb-physics</link><guid isPermaLink="false">2026-11-14/physics</guid><description>Could dark matter and dark energy, the two unseen components that dominate the cosmos, be exchanging energy? This preprint tests a model in which they do so in a way that leaves the expansion of the universe unchanged but speeds up or slows the clumping of dark matter. Data from the Planck satellite alone prefer faster clumping, and allowing it eases a mild disagreement between Planck and the DESI galaxy survey. But faster clumping worsens the disagreement with measurements of how gravity distorts the images of distant galaxies. With those measurements included, the preferred exchange is zero and the standard model of cosmology is favoured. A separate, larger disagreement over the expansion rate is unaffected.</description><pubDate>Sat, 14 Nov 2026 07:07:07 GMT</pubDate><category>Physics</category></item><item><title>Chemistry &amp; Materials Science: A Unified Reactivity Scale Predicts Nonenzymatic Phosphoryl Transfer and Nucleophilic Catalysis in Water</title><link>https://stemsynapse.org/papers/2026-11-14/#sb-chemistry-materials</link><guid isPermaLink="false">2026-11-14/chemistry-materials</guid><description>Living cells constantly move phosphate groups from one molecule to another; it is how they handle energy. Enzymes do this today, so how did such reactions run before enzymes existed? In water, a phosphate donor may pass its phosphate to the intended partner or simply be broken down, and predicting which has been difficult. The researchers measured 11,520 reactions and, with earlier published values, built one scale that ranks 34 donors and 45 acceptors and predicts how fast any pair will react. The scale also predicts which small molecules best speed these reactions up. For donors like ATP, the best is imidazole, the reactive part of the amino acid histidine, which may explain why enzymes use histidine so often.</description><pubDate>Sat, 14 Nov 2026 07:07:07 GMT</pubDate><category>Chemistry &amp; Materials Science</category></item><item><title>Biology &amp; Ecology: Children with developmental language disorder show selective impairments in low-frequency neural tracking of speech</title><link>https://stemsynapse.org/papers/2026-11-14/#sb-biology-ecology</link><guid isPermaLink="false">2026-11-14/biology-ecology</guid><description>Speech has rhythm: slow rises and falls that mark phrasing and stress, and faster beats that mark syllables. One theory holds that children learn language partly because their brain activity locks onto these rhythms. The researchers tested whether that differs in children with developmental language disorder, who struggle with language for no obvious reason. They recorded the brain&apos;s magnetic signals while children with and without the disorder listened to a story. In children with the disorder, brain activity followed the slowest rhythms less closely across hearing and speech areas on both sides, and followed the syllable rhythm less closely on the right. The closer a child&apos;s tracking of slow rhythms, the better their language scores.</description><pubDate>Sat, 14 Nov 2026 07:07:07 GMT</pubDate><category>Biology &amp; Ecology</category></item><item><title>Health &amp; Biomedical: The contribution of Neanderthal ancestry to cancer risk</title><link>https://stemsynapse.org/papers/2026-11-14/#sb-health-biomedical</link><guid isPermaLink="false">2026-11-14/health-biomedical</guid><description>Many people today carry a little DNA inherited from Neanderthals, and some of it affects the immune system and metabolism, both of which bear on cancer. Its effect on inherited cancer risk had not been examined thoroughly. This preprint went through known genetic risk signals for 17 cancers. For twelve, several Neanderthal variants were tied to risk. Closer analysis picked out 32 cases across seven cancers where the Neanderthal variant itself is probably responsible, some raising risk and some lowering it. Overall, Neanderthal DNA accounts for less of inherited cancer risk than expected, a sign that harmful variants were weeded out. Breast cancer is the exception: there it accounts for more, in both East Asian and European data.</description><pubDate>Sat, 14 Nov 2026 07:07:07 GMT</pubDate><category>Health &amp; Biomedical</category></item><item><title>Climate, Energy &amp; Environment: Rapid Emission–Economy Decoupling alongside the Urbanization of Chinese Counties</title><link>https://stemsynapse.org/papers/2026-11-14/#sb-climate-energy</link><guid isPermaLink="false">2026-11-14/climate-energy</guid><description>Nitrogen oxides, pollutants from burning fuel in vehicles, power plants and industry, are a main ingredient of smog. Tracking them place by place across China has been hard because detailed, up-to-date emission records are lacking. The researchers used satellite measurements to estimate summer emissions from 2018 to 2024 for 2,842 counties. Emissions fell in about 84% of them. Of the 1,779 counties with economic data, about 77% cut emissions while their economies grew. The cuts tended to be larger in more urban counties, so a growing share of the country&apos;s emissions now comes from less urban ones, particularly in the west, which have had less attention. The authors suggest pollution control may need to shift its focus there.</description><pubDate>Sat, 14 Nov 2026 07:07:07 GMT</pubDate><category>Climate, Energy &amp; Environment</category></item><item><title>Psychology &amp; Cognitive Science: Prolonged exposure, eye movement desensitization and reprocessing therapy, and cognitive restructuring versus waiting-list for individuals with PTSD and psychosis (RE.PROCESS): A multicenter, four-arm, assessor-masked, superiority randomized controlled trial</title><link>https://stemsynapse.org/papers/2026-11-14/#sb-psychology-cogsci</link><guid isPermaLink="false">2026-11-14/psychology-cogsci</guid><description>Therapies that work through traumatic memories are standard for post-traumatic stress disorder, but evidence has been thin for people who also have a psychotic disorder such as schizophrenia. This trial assigned 162 such patients at random to a waiting list or to one of three therapies, alongside usual care: prolonged exposure, in which the person revisits the memory; EMDR, which pairs recalling it with guided eye movements; and cognitive restructuring, which works on the beliefs around it. Each ran for up to 16 sessions, and symptom raters did not know who had received what. Six months on, all three had reduced PTSD symptoms substantially compared with waiting. Symptoms of psychosis did not change. The authors call for larger trials.</description><pubDate>Sat, 14 Nov 2026 07:07:07 GMT</pubDate><category>Psychology &amp; Cognitive Science</category></item><item><title>Psychology &amp; Cognitive Science: Altered effective connectivity during cognitive distancing in cognitive behavioral therapy nonresponders with depression</title><link>https://stemsynapse.org/papers/2026-11-14/#sb-psychology-cogsci</link><guid isPermaLink="false">2026-11-14/psychology-cogsci/doi:10.1017/s0033291726106138</guid><description>Cognitive behavioural therapy helps only about half of people with major depression. A core skill it teaches is distancing: stepping back from an upsetting thought or image to see it differently. Before 77 patients began 25 sessions of therapy, the researchers scanned their brains while they practised distancing and modelled how brain regions influenced one another. In patients who later improved, frontal regions involved in control influenced each other more strongly. In those who did not, the amygdala, a region involved in emotion, had more influence on the frontal regions. These patterns predicted the outcome better than chance, though only modestly, and added little to what ordinary clinical measures already showed. The authors say the result needs repeating.</description><pubDate>Sat, 14 Nov 2026 07:07:07 GMT</pubDate><category>Psychology &amp; Cognitive Science</category></item><item><title>Math &amp; Computer Science Foundations: Nonlinear lower-tail large deviations at criticality</title><link>https://stemsynapse.org/papers/2026-11-14/#sb-math-cs-foundations</link><guid isPermaLink="false">2026-11-14/math-cs-foundations</guid><description>Draw a network by taking a set of points and joining each pair with a link at random. How likely is it that the result contains no triangle, three points all joined to one another? The answer has long been known when links are rare and when they are common, but not at the awkward middle density where the standard mathematical tools give no sharp answer. This paper develops general methods for such in-between cases and fills the gap. The answer changes character abruptly at one particular density, what physicists call a phase transition. For networks avoiding longer loops with an even number of links, the authors show there is no such jump.</description><pubDate>Sat, 14 Nov 2026 07:07:07 GMT</pubDate><category>Math &amp; Computer Science Foundations</category></item><item><title>AI &amp; Machine Learning: Unbundling Multimodal Vision Encoding</title><link>https://stemsynapse.org/papers/2026-11-13/#sb-ai-ml</link><guid isPermaLink="false">2026-11-13/ai-ml</guid><description>AI models that work with both images and text differ in how they take in a picture. Most pass it through a separate vision component trained beforehand; some train that component together with the language model, and some drop it. Published models also differ in data, design and computing budget, so comparing them cannot show which choice matters. Here every model was trained from scratch on the same data with the same computing budget. A vision component trained jointly did as well as a pre-trained one, so the advance training was not needed. What mattered was having part of the model devoted to vision: models without it fell behind even when enlarged to match.</description><pubDate>Fri, 13 Nov 2026 07:08:57 GMT</pubDate><category>AI &amp; Machine Learning</category></item><item><title>Tech &amp; Engineering: Lazare, veni foras: Reprogramming Dormant Neurons to Backdoor AI Agents</title><link>https://stemsynapse.org/papers/2026-11-13/#sb-tech-engineering</link><guid isPermaLink="false">2026-11-13/tech-engineering</guid><description>AI agents increasingly run for hours unattended with wide access to a computer, so a tampered model is a security risk. This preprint shows how little tampering it takes. The method edits fewer than ten of a model&apos;s unused internal units, with no retraining, so that when a chosen trigger text appears, the model outputs a command picked by the attacker. In one model, changing seven units out of more than 12,000 set off the command in at least 98.8% of test runs containing the trigger, and in none of the normal runs checked. Two other models were just as vulnerable. The edits leave traces: automated defences flagged 91.4% of tampered models, while wrongly flagging 9.7% of clean ones.</description><pubDate>Fri, 13 Nov 2026 07:08:57 GMT</pubDate><category>Tech &amp; Engineering</category></item><item><title>Space, Astronomy &amp; Aerospace: JWST Rocky Worlds DDT Observations of LTT 1445 A b and c in Secondary Eclipse Using MIRI/F1500W</title><link>https://stemsynapse.org/papers/2026-11-13/#sb-space</link><guid isPermaLink="false">2026-11-13/space</guid><description>Do small rocky planets orbiting red dwarfs, the most common kind of star, keep their atmospheres? The James Webb Space Telescope is testing this by measuring the heat such planets give off. This study looked at two rocky planets of LTT 1445 A, the nearest system of its kind, about 22 light-years away. Both gave off as much heat as dark, bare rock would. For planet b, the data limit the surface pressure to about a tenth of Earth&apos;s or less, so it has no substantial atmosphere; planet c is less certain. Planet b sits close to the predicted dividing line between planets that keep their air and those that lose it, so the result helps fix that line.</description><pubDate>Fri, 13 Nov 2026 07:08:57 GMT</pubDate><category>Space, Astronomy &amp; Aerospace</category></item><item><title>Physics: Long-lived light-driven bimagnons at finite momentum in NiO</title><link>https://stemsynapse.org/papers/2026-11-13/#sb-physics</link><guid isPermaLink="false">2026-11-13/physics</guid><description>In a magnet, the tiny magnetic moments of atoms can ripple in waves called magnons. In some materials light can create these ripples in linked pairs, but how long a pair survives was unclear; earlier measurements suggested they fall out of step within 200 femtoseconds, millionths of a billionth of a second. The researchers hit nickel oxide with pulses of infrared light and followed what happened using ultrafast X-ray pulses. The material&apos;s overall magnetic order stayed intact. A signal from the paired ripples built up over about 4 picoseconds, then faded with a characteristic time of 28 picoseconds, more than a hundred times longer than the earlier figure. Such pairs can persist far longer than had been thought.</description><pubDate>Fri, 13 Nov 2026 07:08:57 GMT</pubDate><category>Physics</category></item><item><title>Biology &amp; Ecology: The importance of genes does not predict their mutation rate</title><link>https://stemsynapse.org/papers/2026-11-13/#sb-biology-ecology</link><guid isPermaLink="false">2026-11-13/biology-ecology</guid><description>Mutations are usually treated as random, striking genes without regard to how much they matter. Some researchers have claimed otherwise: that evolution has arranged for important genes to mutate less and for genes needing rapid change to mutate more. Those claims have been disputed. This study reanalysed records of mutations accumulating in normal strains of the gut bacterium E. coli and of baker&apos;s yeast, taking other influences into account. In E. coli, heavily used genes may mutate slightly less, but yeast shows no such pattern. In neither species do essential genes, or genes for abundant proteins, mutate less. Where a gene sits relative to DNA copying made no difference in E. coli either.</description><pubDate>Fri, 13 Nov 2026 07:08:57 GMT</pubDate><category>Biology &amp; Ecology</category></item><item><title>Health &amp; Biomedical: Sub-Threshold Stochastic Vibration Improves Fingertip Tactile Sensitivity in Stroke Survivors and Healthy Adults, with Source-Localized EEG Evidence for Cortical Network Reorganization in Healthy Adults</title><link>https://stemsynapse.org/papers/2026-11-13/#sb-health-biomedical</link><guid isPermaLink="false">2026-11-13/health-biomedical</guid><description>The sense of light touch often fades with age and after a stroke. This preprint tested an odd-sounding remedy: adding random vibration too weak to feel. Nineteen healthy adults and five stroke survivors had a device on one fingertip that delivered a steady vibration, with or without the faint random vibration added, and the researchers measured the weakest steady vibration each person could detect. With the faint random vibration, every participant detected weaker vibrations than before; sensitivity improved by up to 19 times in healthy adults and up to 4 times in stroke survivors. Brain recordings in the healthy group showed stronger coordination between brain regions, especially frontal ones. Only 24 people took part.</description><pubDate>Fri, 13 Nov 2026 07:08:57 GMT</pubDate><category>Health &amp; Biomedical</category></item><item><title>Math &amp; Computer Science Foundations: The Tournament Equilibrium Set Is PSPACE-Complete, Even under Hereditary Uniqueness</title><link>https://stemsynapse.org/papers/2026-11-13/#sb-math-cs-foundations</link><guid isPermaLink="false">2026-11-13/math-cs-foundations</guid><description>Some voting rules pick winners from head-to-head contests: for every pair of candidates, which would a majority prefer? One such rule, proposed by Schwartz in 1990, judges each candidate by the candidates that beat it, in a definition that refers back to itself. Deciding whether a given candidate is among its winners was known to be hard to compute, but not exactly how hard. This paper settles it: the problem belongs to a class believed to be far beyond efficient computation, and stays that hard even under a condition Schwartz conjectured always holds. Yet in the very cases built to be hard, one winner can be found quickly. Finding some winner is easy; confirming a particular one is not.</description><pubDate>Fri, 13 Nov 2026 07:08:57 GMT</pubDate><category>Math &amp; Computer Science Foundations</category></item><item><title>Space, Astronomy &amp; Aerospace: Foiling Black Hole Foils: Revealing Horizon Alternatives with Baryonic Atmospheres</title><link>https://stemsynapse.org/papers/2026-11-12/#sb-space</link><guid isPermaLink="false">2026-11-12/space</guid><description>A black hole has an event horizon, a boundary nothing can return from. That boundary is hard to square with quantum physics, so theorists have proposed look-alikes with a surface instead. Could astronomers tell the difference? The authors worked out what happens to gas falling steadily onto such an object. In typical cases the gas should build into a dense atmosphere that shines, with a brightness set mainly by how fast gas falls in and only weakly by the object underneath. So where matter falls onto a black hole candidate and no such glow is seen, broad classes of these look-alikes are constrained or ruled out, provided gravity outside behaves as Einstein&apos;s theory says.</description><pubDate>Thu, 12 Nov 2026 22:30:39 GMT</pubDate><category>Space, Astronomy &amp; Aerospace</category></item><item><title>Chemistry &amp; Materials Science: MgO Facets as Regenerable Molecular Oxygen Reservoirs in Oxidative Coupling of Methane</title><link>https://stemsynapse.org/papers/2026-11-12/#sb-chemistry-materials</link><guid isPermaLink="false">2026-11-12/chemistry-materials</guid><description>Catalysts speed up chemical reactions at their surfaces, and the same material can behave differently depending on which crystal faces it shows. The researchers compared magnesium oxide particles with different exposed faces in a test reaction that joins methane molecules into larger ones. Eight-sided particles, which expose one particular face, consistently outperformed the conventional forms. Measurements taken during the reaction at about 800 °C, with calculations, indicate why: oxygen molecules held on that face help break methane&apos;s carbon-hydrogen bonds. The oxygen is used up as the reaction runs, but exposing the catalyst to oxygen restores it, so the cycle can be repeated. Shaping a metal oxide&apos;s surface can therefore give it an ability the ordinary form lacks.</description><pubDate>Thu, 12 Nov 2026 22:30:39 GMT</pubDate><category>Chemistry &amp; Materials Science</category></item><item><title>Biology &amp; Ecology: Life-cycle plasticity enables conditional asexual reproduction in a kelp</title><link>https://stemsynapse.org/papers/2026-11-12/#sb-biology-ecology</link><guid isPermaLink="false">2026-11-12/biology-ecology</guid><description>Many living things can reproduce either sexually or without a partner, but how they switch between the two is poorly understood. The researchers studied this in a kelp, Laminaria pallida, a large seaweed with a complex life cycle, using measurements of gene activity, growth experiments and samples from wild populations. Reproducing without a partner was linked to lower fitness. In two wild populations the ability was kept but rarely used. And when potential mates were present, the kelp developed asexually less often. The authors suggest that asexual reproduction is a fallback, letting the kelp persist when sex is not possible while keeping the long-term benefits of sex when mates are around.</description><pubDate>Thu, 12 Nov 2026 22:30:39 GMT</pubDate><category>Biology &amp; Ecology</category></item><item><title>Health &amp; Biomedical: Effect of intensive blood pressure control on cognitive outcomes and brain structures in adults with type 2 diabetes: prespecified analysis of secondary outcomes of BPROAD randomised controlled trial</title><link>https://stemsynapse.org/papers/2026-11-12/#sb-health-biomedical</link><guid isPermaLink="false">2026-11-12/health-biomedical</guid><description>Would lowering blood pressure further than usual protect the brain as well as the heart? This trial tested the idea in adults with type 2 diabetes. At 145 centres in China, 12,434 people aged 50 or older, none with probable dementia, were assigned to aim for a systolic pressure, the upper number in a reading, either below 120 or below the standard 140, for up to five years. Over a median of 3.8 years, dementia or mild cognitive impairment developed in 747 people in the intensive group and 783 in the standard group, a difference small enough to be due to chance. Aiming lower did not clearly protect thinking and memory in this group.</description><pubDate>Thu, 12 Nov 2026 22:30:39 GMT</pubDate><category>Health &amp; Biomedical</category></item><item><title>Climate, Energy &amp; Environment: Mineral‐Associated Carbon Accrual Masks Labile Particulate Carbon Loss in Permafrost Region on the Tibetan Plateau</title><link>https://stemsynapse.org/papers/2026-11-12/#sb-climate-energy</link><guid isPermaLink="false">2026-11-12/climate-energy</guid><description>Frozen soils hold vast amounts of carbon, and what becomes of it as the climate warms is uncertain. Soil carbon comes in different forms: some is bound to minerals and lasts, some is loose plant debris that microbes break down easily. Using measurements from 104 sites across the Tibetan permafrost region and a computer model, the authors projected each form through this century under three emissions scenarios. Mineral-bound carbon, already 56.5% of the total, kept building up, so total soil carbon rose in every scenario. But under the highest emissions the light, easily decomposed carbon declined, most sharply 10 to 30 cm down. A net gain can therefore hide the loss of the carbon that soil microbes depend on.</description><pubDate>Thu, 12 Nov 2026 22:30:39 GMT</pubDate><category>Climate, Energy &amp; Environment</category></item><item><title>Math &amp; Computer Science Foundations: Reconciling Discrete-Time Mixed Policies and Continuous-Time Relaxed Controls in Reinforcement Learning and Stochastic Control</title><link>https://stemsynapse.org/papers/2026-11-12/#sb-math-cs-foundations</link><guid isPermaLink="false">2026-11-12/math-cs-foundations</guid><description>Reinforcement learning trains a system by trial and error, and it is usually described in steps: at each tick of a clock the system picks an action, often at random according to a rule called a mixed policy. Some researchers have recently recast it in continuous time, where the state changes smoothly and randomly, and have treated the two descriptions as fully equivalent. This paper asks how far that holds. For the case where the controller influences only the predictable trend of the system, and not its random fluctuations, the authors prove that the step-by-step method converges to the continuous one as the time steps shrink. Where control also affects the randomness, they set out the technical obstacles that remain.</description><pubDate>Thu, 12 Nov 2026 22:30:39 GMT</pubDate><category>Math &amp; Computer Science Foundations</category></item></channel></rss>