Julius Adolff, Uroš Seljak
AstronomerThe authors develop a significantly faster field-level likelihood as an alternative to optimal power-spectrum analysis requiring costly covariance-matrix inversion. On sufficiently large scales, the galaxy overdensity field is approximately Gaussian, making the power spectrum an approximately lossless compression, but survey-window-induced mode coupling renders standard estimators suboptimal. For local-type primordial non-Gaussianity, the field-level likelihood forecasts a 14% improvement in constraining power over standard power-spectrum analysis. Applied in forecasts for the DESI DR2 LRG and QSO samples, it yields σ(f_NL) = 4.1 versus 4.8 for the power-spectrum approach; these are projected uncertainties, not measured constraints. The authors also present a non-parametric method for mitigating angular systematics through efficient deprojection of contaminated modes.
PublicThe way a galaxy survey covers the sky can mix together patterns in galaxy clustering, making standard analyses less effective on the largest scales. The authors develop a faster approach that works directly with the galaxy density field. They forecast that it could constrain a possible departure from Gaussian patterns in the early universe 14% better than a standard power-spectrum analysis. For the DESI DR2 galaxy and quasar samples considered, the projected uncertainty on the parameter f_NL falls from 4.8 to 4.1. This is a forecast of the method’s sensitivity, not a detection of such an early-universe signal. The authors also propose a flexible way to reduce unwanted patterns across the sky that could contaminate the analysis.
Young explorerScientists look at patterns in where galaxies are to learn about how the universe began. But the shape of the area they can see makes some patterns harder to untangle. The researchers made a faster way to study them. Their calculations suggest it could give more precise clues about the early universe, but that is a prediction, not a discovery. They also suggest a way to reduce unwanted patterns that could confuse the clues.
Selim C. Hotinli, Edmond Chaussidon, Simone Ferraro+1 more
AstronomerThe authors report the first South Pole Telescope reconstruction of the large-scale radial velocity field using the kinetic Sunyaev–Zel’dovich effect. They cross-correlate SPT-3G D1 maps at 95, 150, and 220 GHz with luminous red galaxies from the DESI Legacy Imaging Surveys. Combining frequencies for the extended galaxy sample yields galaxy–velocity correlation and velocity power spectrum detections at signal-to-noise ratios of 11.2 and 8.1, respectively. Reconstructed velocities are detected at 220 GHz for the first time, with signal-to-noise ratios of 4.2 and 3.1 for the correlation dipole and monopole, and 3.8 for the velocity power spectrum. Foreground modeling separately detects thermal Sunyaev–Zel’dovich and cosmic infrared background contributions to the monopole. Across the approximately 5% sky overlap, reconstructed velocities constrain local-type primordial non-Gaussianity to f_NL = -38 (+61/-61); adding galaxy clustering gives -20 (+40/-46). Both constraints are consistent with zero. The galaxy–velocity correlation is consistent across all three frequencies.
PublicResearchers have used South Pole Telescope data to reconstruct large-scale motions toward or away from us through the kinetic Sunyaev–Zel’dovich effect. They combined maps at three microwave frequencies with a catalogue of galaxies, detecting both a relationship between galaxies and the reconstructed motions and a statistical pattern in those motions. This is the first such reconstruction with the South Pole Telescope, and the first detection of reconstructed velocities at 220 GHz. The galaxy–motion relationship agrees across the three frequencies. The team also modeled foreground signals and separately detected two contributions that affect part of the analysis. Using about 5% of the sky, they placed new limits on possible departures from Gaussian statistics in the early universe. Adding information about galaxy clustering tightened those limits, but the results remain consistent with no such departure.
Young explorerScientists used a telescope at the South Pole and a big list of galaxies to work out patterns of things in space moving toward us or away from us. Three kinds of telescope maps gave matching clues. The patterns also helped test ideas about the early universe, but they did not show a clear sign of the special pattern being tested.
Michael Kovač, Andrina Nicola, Jozef Bucko+8 more
AstronomerThis simulation-based study develops two ways to generate Compton-y maps from dark-matter-only N-body light cones using component-wise baryonification, aiming to calibrate baryonic feedback through joint weak-lensing and thermal Sunyaev-Zel'dovich (tSZ) inference. The model separates gas and stellar fields and derives gas pressure from hydrostatic equilibrium. Painting halo pressure profiles and distributing the signal along baryonified gas agree within a few per cent. Calibration jointly fits gas mass fractions and thermal pressure profiles in the fiducial and strongest-AGN-feedback FLAMINGO simulations, with non-thermal pressure measured from simulated gas velocities rather than fitted freely. Predicted Compton-y profiles agree within halo-to-halo scatter; painting onto FLAMINGO halos reproduces both models' tSZ power spectra to about 5% over 500 ≲ ℓ ≲ 3000. Maps from a CosmoGridV1-like simulation agree with FLAMINGO to about 10% over 300 ≲ ℓ ≲ 8000. Residual differences arise from the few hundred most massive clusters, and removing these from both catalogs brings agreement. These simulation tests support efficient tSZ forward modeling toward, rather than demonstrate, joint cosmological inference for Euclid, LSST, and the Simons Observatory.
PublicHot gas moved around by processes such as feedback from active galactic nuclei complicates efforts to study the universe with weak gravitational lensing. That gas also leaves a thermal Sunyaev-Zel'dovich (tSZ) signal, so modeling the two together could help reduce this uncertainty. The researchers developed two methods for adding simulated gas signals to simulations that initially follow only dark matter. Both use a model that separates gas and stars and calculates gas pressure; the methods agree within a few per cent. They calibrated and tested the model against two FLAMINGO simulations with different feedback strengths, not against sky observations. When applied to FLAMINGO's halos, the model reproduced the strength of tSZ fluctuations to about 5% over the tested range of angular scales. In a separate, CosmoGridV1-like simulation, agreement was about 10% over a broader range. The remaining differences came from the few hundred most massive galaxy clusters; removing those clusters from both catalogs brought the predictions into agreement. The work is a step toward combining lensing and hot-gas signals to improve future cosmological studies, rather than a new measurement of the universe.
Young explorerScientists make pretend universes on computers to help understand the real one. But hot gas can move around, and that makes the pretend universes tricky to build. The scientists tried two ways to add the gas's signal to computer universes that started with only dark matter. The two ways gave nearly the same answer and were close to more detailed computer models. The biggest groups of galaxies caused some differences. This could help scientists study the universe later, but this test used computers, not new pictures of the sky.
Sambit K. Giri, P. Daniel Meerburg, Léon V. E. Koopmans
AstronomerThe authors develop the phase gradient spectrum as an efficient probe of non-Gaussianity in the reionization 21-cm signal, complementing the phase-blind power spectrum. It measures the circular coherence of phase differences between Fourier modes separated by a long-wavelength mode and is shown analytically to measure the angle-averaged normalised squeezed bispectrum, linking small-scale power to long-wavelength fluctuations. In simulations spanning different ionizing-source properties and simulation frameworks, it evolves non-monotonically: initially tracing matter-density non-Gaussianity, then ionization topology once the mean neutral fraction falls below approximately 0.9. Phase coherence reaches a minimum around the midpoint of reionization and increases as neutral regions become isolated islands. Sign-preserving statistics identify the minimum with a squeezed-bispectrum sign change. These are simulation and analytic results, not an observational detection. A Fisher forecast combining five redshifts predicts astrophysical parameter constraints from this statistic alone for a 1000-hour SKA-Low observation, tightening by a factor of at least approximately two for 10,000 hours. Combining it with the power spectrum improves marginalised constraints by up to approximately 20% relative to the power spectrum alone.
PublicThe researchers propose a way to extract more information from the hydrogen radio signal associated with cosmic reionization. Standard power-spectrum analysis misses how the phases of the signal's wave components line up. Their new measure captures some of that information more efficiently than a commonly used higher-order statistic. In computer simulations, the measure first reflects the matter distribution, then the pattern of ionized and neutral regions. Its phase coherence falls to a minimum around halfway through reionization, then rises as the remaining neutral hydrogen shrinks into isolated patches. This behavior appears across several models and simulation approaches; it has not been reported here as an observed signal. Forecasts for future SKA-Low observations suggest that adding the new measure to the power spectrum could tighten constraints on the astrophysical parameters by up to about 20%.
Young explorerLong ago, hydrogen gas in space was changing, leaving smaller patches of unchanged gas. Scientists made computer versions of this story and found a new way to study patterns in the gas's radio waves. Their pattern measure went down and then up as the patches became isolated. They predict this method could help a telescope learn more, but they have not yet shown it working on real sky observations.
Azadeh Moradinezhad Dizgah, Zucheng Gao, Tom Chiesura
AstronomerThis halo-model forecast assesses cross-correlations of SPHEREx near-infrared EBL maps with CMB lensing at Planck and Simons Observatory (SO) sensitivity, jointly fitting integrated galaxy light (IGL), intrahalo light (IHL), and emission lines. Without radial filtering, Planck-level lensing yields a forecast 9σ IGL detection in the deep fields but no line detection. For the all-sky survey, forecast IGL and IHL amplitude uncertainties are 1.0% and 3%, with a 4.6σ detection of five lines fitted as a single template; SO-level lensing improves these to 0.6%, 1.6%, and 7.4σ. Removing radial modes longer than about 600 Mpc reduces every component's signal-to-noise ratio by at least 90% in the joint fits. Continuum modeling, rather than filtering, is therefore central to line recovery. The quoted forecasts assume a known continuum spectral shape: calibration using continuum auto-spectra retains 7.3–7.4σ line significance, whereas constraining the shape from the cross-spectrum alone leaves 1.4–2.3σ. Line detection significances vary by factors of seven to eleven across line models. These are forecasts, not measured detections, and support IGL as a robust target while making line recovery conditional on continuum modeling across channels.
PublicThe near-infrared sky contains distinctive emission lines from galaxies, buried beneath much brighter, smoothly varying light. This study forecasts how well researchers could separate those contributions by comparing SPHEREx maps with maps of how matter bends the cosmic microwave background. That comparison rejects foreground light unrelated to the matter, but the bright extragalactic continuum remains a challenge because it also traces matter. For the all-sky survey, the forecasts suggest precise measurements of light from galaxies and their surrounding halos, plus a possible detection of five emission lines treated together. The combined line significance reaches 4.6σ with Planck-level lensing maps or 7.4σ with Simons Observatory-level maps under favorable continuum assumptions. Filtering out the smooth light also removes most of the useful shared signal; modeling it is more promising. The line prospects depend strongly on the assumed line model and on how well the continuum's spectrum can be established. These are predictions of what the analysis could achieve, not reported observations.
Young explorerLight from faraway galaxies is like a faint tune hidden under a much louder hum. Scientists calculated how they might find that tune by comparing a map of the light with a map showing how matter bends very old light. Their calculations suggest that erasing the hum would erase too much of the tune, too. Understanding the hum and including it in their calculations could work better. They have not found the hidden tune this way yet, and how well it works depends on getting the hum right.
Azadeh Moradinezhad Dizgah, Zucheng Gao
AstronomerThis simulation study models cross-correlations of near-infrared intensity with CMB-lensing convergence, which reject foregrounds and noise uncorrelated with dark matter. Simulated SPHEREx maps include integrated galaxy light (IGL), intrahalo light (IHL), and emission lines on a common Hidden Valley lightcone covering 100 square degrees to z ≈ 6. Stellar-population synthesis assigns halo emission using either fixed or stellar-mass-dependent metallicity. Halo-model predictions using simulation-derived luminosity–halo mass relations and measured clustering, without free parameters, reproduce IGL and line auto-spectra and lensing cross-spectra across the band; at 2.2 micrometres, agreement is within 5% above ℓ ≈ 400 for auto-spectra and 7% above ℓ = 200 for cross-spectra. Nonlinear clustering is essential: linear theory describes a line–lensing cross-spectrum up to ℓ ≈ 1000–1600 at that wavelength, but fails for IGL and convergence at multipoles of a few hundred. Assuming SPHEREx deep-field and Simons Observatory lensing noise, forecast unfiltered signal-to-noise ratios, treating each component alone, are 9–21 for lines, 20 for IGL, and 15 for IHL. Removing radial modes longer than about 600 Mpc retains 62% of IGL signal-to-noise but at most 20% for lines. Fitted amplitudes remain within about 1σ of predictions before and after filtering. Emission-model choices change channel intensities by up to threefold, but line detectability by only tens of percent. These are simulation tests and forecasts, not observational detections.
PublicThe researchers used computer simulations to explore how infrared maps could be compared with maps of how matter bends light from the early universe. This comparison helps reject foreground light and noise that do not follow dark matter. Their simulated infrared maps contain light from galaxies, diffuse light within their surrounding halos, and light emitted at particular wavelengths. A model that includes nonlinear matter clustering closely reproduces several of the simulated signals; a simpler linear description is insufficient for some of them. With assumed noise levels for SPHEREx and Simons Observatory, the study forecasts detectable correlations when each light component is considered separately—not detections in real observations or a demonstration that all components can be separated together. A filter that removes variations over long distances along the viewing direction reduces the predicted significance of emission-line signals much more than that of galaxy light. Different assumptions about the emitting populations strongly change predicted brightness, but change emission-line detectability more modestly.
Young explorerScientists made pretend sky maps on a computer. They put in light from galaxies and compared it with a map of how matter bends very old light. Matching the maps could help them ignore some unwanted light and noise. Their computer tests suggest this could work with planned measurements, but they have not found these signals in real sky pictures here.
Georgios N. Vassilakis, Miles Cranmer, Vasily Belokurov
AstronomerSpectro-Spatial Survey Fusion jointly fits imaging and spectroscopy through each instrument’s response using a continuous spectral surface-brightness field. Adapted from Gaussian splatting, the model combines adaptive spatial Gaussians whose spectra use a shared, learned template dictionary. Demonstrated on 400 SDSS ugriz exposures and 800 fiber spectra over an 84-by-84-arcminute COSMOS region, it reproduces broad spatial and spectral diversity among galaxies, stars, and quasars. For fitted observations, summed predicted flux in source-selected pixels agrees within 2 percent in every band, and 96 percent of fitted spectra agree within 10 percent in integrated flux. The shared dictionary permits model spectra to be queried throughout the imaged area, including locations without fitted spectroscopy; these are model estimates, not new spectral measurements. Without refitting, the model also predicts withheld SDSS data and selected HSC images and DESI spectra excluded from fitting, recovering broad spiral-arm patterns in two selected galaxies and continuum shapes for galaxies without fitted fibers. These demonstrations support a route toward joint survey modeling, rather than establishing uniformly accurate spectral reconstruction everywhere.
PublicA new framework combines sky pictures and spectra—the measurements that separate light into its different wavelengths—into one model, rather than analyzing them separately. Called Spectro-Spatial Survey Fusion, it was tested using 400 Sloan Digital Sky Survey images and 800 spectra from a region called COSMOS. For the data used to build the model, the total predicted light in selected source pixels matched the images within 2 percent in each of five wavelength bands, while 96 percent of the spectra matched within 10 percent in total light. The researchers also tested predictions against observations left out of the fit, including selected images and spectra from other surveys. Without further fitting, the model predicted broad spiral-arm patterns in two selected galaxies and the overall spectral shapes of galaxies lacking fitted spectra. It can estimate a spectrum anywhere in the pictured region, but those estimates are not direct measurements. The work demonstrates a way to let different surveys inform a common model of the sky.
Young explorerPictures show where things are in the sky. Another kind of observation spreads their light into colors. Scientists made a computer model that learns from both kinds together. It matched much of the information they gave it. They also checked it with observations it had not learned from: it predicted the big spiral shapes in pictures of two galaxies and the general mix of light colors from some other galaxies. The model can guess the light colors in places where it only learned from pictures, but a guess is not the same as measuring them.
Shane Brady, Eoin Ó Colgáin, M. M. Sheikh-Jabbari
AstronomerThe authors revisit the decreasing inferred H0 trend with lens redshift reported by H0LiCOW/TDCOSMO in 2019 within flat ΛCDM at less than 2σ significance. Using TDCOSMO 2026 results strengthens the trend in a like-for-like comparison, but propagating mass-sheet transformation uncertainties reduces its significance to approximately 1σ. The Akaike Information Criterion currently favours constant H0 over a decreasing trend by 2.4:1. The authors note agreement with negative slopes from independent observables and forecast that approximately 40 and 90 lenses could establish the trend at approximately 2σ and 3σ, respectively, conditional on the feature being physical and current uncertainties being representative. Such a trend could indicate missing late-universe physics, but is not established. They interpret the TDCOSMO 2026 constraint of H0 ≈ 71 km/s/Mpc as an average over the lens redshift distribution, naturally anchored at intermediate lens redshifts.
PublicA new analysis revisits hints that estimates of the universe’s expansion rate, called the Hubble constant, decrease with the redshift of the gravitational lenses used to measure it. Although the hint looks stronger than in earlier comparable analyses, including uncertainty in the lenses’ mass models leaves only weak evidence, at about one standard deviation. A statistical model comparison currently favours a constant value over a decreasing trend by 2.4 to 1. The authors report agreement with downward trends in independent measurements, but this does not establish that the effect is real or that the standard cosmological model is missing physics. They forecast that about 40 lenses could yield two-standard-deviation evidence and about 90 could yield three, if the trend is physical and today’s uncertainties are representative. They also interpret the reported expansion-rate estimate of about 71 km/s/Mpc as an average across the lens sample, anchored at intermediate redshifts.
Young explorerScientists use giant space magnifying glasses, called gravitational lenses, to help measure how fast the universe is growing. Measurements from lenses at different distances hint at different answers. But when the scientists include uncertainty about the lenses, the hint is very weak. For now, one shared answer fits better than a changing answer. More lenses might help them find out whether the hint is real.
Wen-Qing Guo
AstronomerThis theoretical study generalizes the halo model for cross-correlations between the unresolved gamma-ray background (UGRB) and large-scale structure to velocity-dependent WIMP annihilation. It considers sigma v_rel proportional to v_rel^n for Sommerfeld-like (n = -1), s-wave (n = 0), p-wave (n = 2), and d-wave (n = 4) scenarios. Within the model, velocity dependence changes both the annihilation emission and the halo populations dominating it, producing distinct effective halo biases, redshift distributions, and angular clustering. The predicted cross-correlations therefore contain information not captured by the overall gamma-ray intensity alone. A statistical forecast using the Fermi LAT UGRB and 2MRS galaxy catalog estimates projected sensitivities to the velocity-independent coefficient in each annihilation prescription. These are model predictions and sensitivity forecasts, not an observational detection of dark matter annihilation.
PublicThe researchers model a possible way to investigate dark matter by comparing the unresolved gamma-ray background with the distribution of galaxies. If dark matter consists of WIMPs that annihilate, the rate of those reactions could depend on how fast the particles move relative to one another. The study explores four possible speed dependences and predicts that they would change which dark matter halos contribute most strongly and how the gamma-ray pattern lines up with cosmic structure. That pattern could help distinguish possibilities that total gamma-ray brightness alone cannot. Using Fermi LAT gamma-ray observations and the 2MRS galaxy catalog as a forecast example, the authors estimate how sensitive such a comparison could be. They do not report detecting dark matter annihilation.
Young explorerDark matter is something scientists are trying to understand. This study asks what might happen if dark matter particles meet and make gamma rays, a kind of light. The scientists used a model to predict how different particle speeds could change the light’s pattern near groups of galaxies. Comparing those patterns might give clues, but this study did not find dark matter making light.
Camille Bonvin
AstronomerThe authors argue that current tests cannot uniquely attribute a deviation from ΛCDM to modified gravity, evolving or clustering dark energy, or non-cold dark matter properties. Such tests depend on unobservable quantities connected to observations through assumed equations or compared with predictions within restricted theory classes, compromising model independence and producing cross-sector degeneracies. They construct three observable-only tests that, in combination, are intended to distinguish the sectors. The key additional information comes from the 0i Einstein equation, alongside the gravitational slip and Euler equations. This is a methodological result, not an observational identification of new physics; its motivation depends on whether the reported cosmological tensions persist.
PublicSome cosmological observations have begun to challenge the standard model of the universe, but it remains uncertain whether these tensions will persist. If they do, possible explanations include changes to gravity, dark energy that changes or clusters, or dark matter with properties beyond the standard cold-dark-matter picture. The authors argue that current tests cannot reliably tell these explanations apart because they rely on quantities that cannot be observed directly and on assumptions used to interpret them. They develop three tests based only on observable quantities that, when combined, could separate these possibilities. A usually untested part of Einstein’s equations supplies crucial extra information. The paper presents a way to investigate the cause, not evidence that any particular explanation is correct.
Young explorerSome things astronomers see in space may not fit their usual picture of how the universe works. If the puzzle stays, it might mean gravity works differently, or that mysterious things called dark energy or dark matter behave differently than expected. The researchers say today’s checks can mix up these answers. They have made three new checks that use only things we can observe and are designed to work together to tell the answers apart. They have not yet shown which answer is right.
André V. Giannini, Victor P. Gonçalves, Samuel M. Sanches Jr
AstronomerThis modeling study follows early-Universe thermodynamics across the QCD crossover and up to the electroweak scale. Using HotQCD as a baseline, the authors generate 1000 thermodynamically consistent equation-of-state samples with Gaussian Process regression, extrapolate them with and without electroweak contributions, and solve the Friedmann equations to predict expansion histories and gravitational-wave spectra. Within these calculations, electroweak contributions dominate over differences among QCD baselines sharing the same lattice calibration. They bring the equation of state toward the radiation-dominated limit earlier, reduce sound-speed excursions toward the causality bound, suppress transient Hubble and kinematic fluctuations near 10^-5 s, and change the predicted gravitational-wave spectrum from strongly frequency-dependent to a flat, suppressed plateau. The results argue that electroweak modeling is essential for unbiased inference of QCD-era physics from a stochastic gravitational-wave background; they are theoretical predictions, not gravitational-wave detections.
PublicThe researchers calculated how the hot early Universe expanded and how its changing contents could affect a background of gravitational waves. They tested 1000 internally consistent versions of a model describing the relationship between pressure, energy density, and temperature, extending each version with and without electroweak contributions. In these calculations, including the electroweak sector mattered more than the differences among the tested strong-interaction models, which shared the same underlying calibration. It smoothed brief changes in the expansion and turned the predicted gravitational-wave spectrum from one that varies strongly with frequency into a lower, flat plateau. This is a modeling result, not an observed signal. It suggests that interpreting such gravitational waves as clues to the strong-interaction era requires careful treatment of electroweak physics too.
Young explorerScientists used computer calculations to explore how the very hot baby Universe grew. They tried many versions of their model and found that including a part of physics called the electroweak sector made a big difference. It made the predicted pattern of tiny ripples in space flatter and lower. They have not seen these ripples in this study; the calculations help explain what scientists might learn from them.
Margherita Lembo, Luca Caloni, Martina Gerbino+5 more
AstronomerThe study updates CMB polarization constraints on renormalizable photon-sector Lorentz violation in the minimal Standard Model Extension, considering CPT-odd operators that source cosmic birefringence and CPT-even operators that source circular polarization. Combining Planck, ACT and SPT temperature and E-mode data with BICEP/Keck, SPT-3G D1 and SPTpol B-mode measurements gives an absolute value of k^(3)_(V)00 below 7.75 × 10^-45 GeV and a magnitude of the vector k_AF below 3.82 × 10^-45 GeV, both at 95% credible level. The authors report these as the strongest bounds to date from any probe, driven by SPT-3G D1 B-mode data. Including the Planck PR4 EB spectrum for the first time in this type of analysis yields a mild preference for a nonzero absolute value of k^(3)_(V)00, not a firm detection. The CPT-even constraint is k_F,E+B below 5.99 × 10^-32; adding the latest CLASS circular-polarization measurements tightens the bound by about 7%, marking the first contribution of CMB circular-polarization data to constraining this coefficient. Forecasts using linear polarization alone suggest SO could tighten the CPT-even bound by about 30%, while LiteBIRD could tighten the CPT-odd bounds by 40–45%. Exploiting circular-polarization sensitivity from its half-wave-plate non-idealities, LiteBIRD could improve the CPT-even bound by more than an order of magnitude; these improvements are forecasts, not measured results.
PublicResearchers used the polarization of the cosmic microwave background, the ancient light from the early universe, to test possible departures from Lorentz symmetry in a particular theory of electromagnetism. Combining several telescopes' measurements, they obtained what they report are the strongest limits yet from any probe on two quantities associated with a rotation of polarization: below 7.75 × 10^-45 GeV and 3.82 × 10^-45 GeV at a 95% credible level. Measurements from SPT-3G D1 drive these limits. Adding a Planck measurement of correlations between two polarization patterns produces only a mild preference for one quantity being nonzero, not a firm discovery. A separate quantity associated with circular polarization is limited to below 5.99 × 10^-32. Adding CLASS circular-polarization measurements tightens that limit by about 7%, the first time such CMB measurements have helped constrain this quantity. Looking ahead, the authors forecast that linear-polarization measurements from the Simons Observatory could tighten the circular-polarization-related limit by about 30%, while LiteBIRD could tighten the other two limits by 40–45%. If LiteBIRD exploits circular-polarization sensitivity arising from imperfections in its half-wave plate, it could improve the first of these limits by more than a factor of ten. These future gains are predictions, not observations.
Young explorerScientists studied very old light from when the universe was young. Light can have different patterns of wiggling, and the scientists checked whether those patterns show tiny changes allowed by an idea they are testing. Their measurements put very strict limits on how big some changes could be. One part of the study gave a small hint of a change, but it was not enough to say they had found one. Another kind of light measurement helped make one limit a little stricter. The scientists also worked out how future experiments might do better, but those better measurements have not happened yet.
Philip F. Hopkins
AstronomerThe authors introduce and numerically test a Lagrangian phase-space method for solving the Vlasov/Boltzmann equation across arbitrary collisionality. Mesh-generating points define spatial volumes and carry local, anisotropic momentum-distribution basis functions. Higher-order reconstruction and Godunov-type integration evolve 10 or 13 moments per basis while explicitly conserving mass, momentum, and energy. The method is designed to represent multistream, collisional, phase-mixed, and thermal distribution functions, with computational cost described as a modest multiple of traditional Monte Carlo, particle-in-cell, or N-body methods. A public GIZMO implementation uses a spatially Lagrangian partition, second-order reconstruction, and bases that maintain constant cell masses. Numerical test problems demonstrate conservation, Galilean, translation, and rotation invariance, and complex distribution-function evolution without spurious dissipation or instability in those tests. Potential advantages for dark matter, stellar dynamics, gas, dust, radiation, cosmic rays, and plasmas are discussed rather than established across all these applications.
PublicThis paper presents a new computer-simulation method for tracking how matter moves, from systems where particles rarely collide to those where collisions are important. Rather than using each simulated particle only as a single sample of motion, the method uses moving regions that each describe a range of particle velocities. This lets it represent both overlapping streams and more thoroughly mixed material while conserving mass, momentum, and energy. In numerical tests, the method handled complicated motion without artificial smoothing or instability and behaved consistently when the system's overall motion, position, or orientation was changed. The authors describe its computing cost as a modest multiple of common particle-based methods and provide an implementation in the public GIZMO code. They discuss possible uses for simulating stars, dark matter, gas, dust, radiation, and plasmas; these are potential applications, not new astronomical observations.
Young explorerScientists made a new way for a computer to follow lots of tiny things moving around. Imagine moving boxes, each keeping track of things going in different directions and at different speeds. The method can handle things that bump into each other often or hardly ever. In computer tests, it kept track of the total amount of stuff, its motion, and its energy without adding unwanted smoothing or making the calculation go unstable. The scientists think it could help make computer models of stars, gas, dust, and other things in space. They tested a computer tool, not something newly seen in the sky.
Eleni Tsaprazi, Federico Marulli, Pier-Stefano Corasaniti+2 more
AstronomerThe authors analyse redshift-space clustering in 85,818 spectroscopic galaxy clusters selected from a catalogue constructed from the DESI Legacy Imaging Surveys, spanning 0.1 ≤ z ≤ 0.4 and M_500c ≳ 6 × 10^13 solar masses. They develop and validate random-catalogue construction and effective halo-bias calibration methods, then fit the two-point correlation function monopole and quadrupole over 10–150 h^-1 Mpc to jointly constrain growth and Alcock–Paczynski (AP) distortions. With the matter power-spectrum shape fixed to a fiducial cosmology, they infer fσ8 = 0.45 ± 0.08 at an effective redshift of 0.28. Excluding baryon acoustic oscillation scales and imposing tight AP priors changes the constraint to 0.52 ± 0.03. Both estimates are consistent with general relativity within the adopted fiducial cosmology; cluster clustering provides a growth probe complementary to cluster abundance.
PublicSome proposed alternatives to Einstein’s general relativity aim to explain dark energy and dark matter. This study tests gravitational growth using the spatial clustering of 85,818 galaxy clusters. The researchers infer a quantity combining how quickly cosmic structure grows with how strongly matter is clumped, finding 0.45 ± 0.08. A different analysis, which excludes the scales associated with baryon acoustic oscillations and tightly constrains geometric distortions, gives 0.52 ± 0.03. Both results agree with general relativity within the cosmological model assumed in the analysis. This is a consistency test, not proof that Einstein’s theory is the only possible explanation.
Young explorerGalaxies can gather in huge groups called clusters. Scientists studied how these clusters are spread out to learn how gravity helps things gather together in space. Their results fit Einstein’s ideas about gravity when they used their chosen model of the universe. That is a useful check, but it does not prove that every other idea is wrong.
Anjan A Sen, Purba Mukherjee
AstronomerThe authors derive a theoretical consistency relation for spatially flat ΛCDM between the acceleration-transition redshift z_t and the redshift z_pk at which the linear-growth observable fσ8 turns over: (1+z_t)/(1+z_pk) = 1.1596. This ratio is independent of the present matter fraction and the overall expansion and growth normalisations, linking background expansion to structure growth. The abstract also reports establishing conditions for extending the relation to broader dark-energy and modified-gravity theories, without specifying those conditions. No observational validation is reported.
PublicThe study identifies a mathematical link between two milestones in a spatially flat version of the standard cosmological model: when cosmic expansion begins accelerating and when a quantity tracking the growth of cosmic structure reaches its peak. The predicted ratio of one plus their respective redshifts is 1.1596, regardless of the model’s present-day matter fraction or its overall expansion and growth scales. This offers a consistency check connecting expansion with structure growth, rather than reporting an observational confirmation. The authors also establish conditions for extending the link to other theories, though the abstract does not detail them.
Young explorerThe universe is getting bigger, and matter gathers into clumps. Using a mathematical model, scientists found a rule linking when the universe starts getting bigger faster with when a number describing how its clumps grow reaches its highest point. The rule could help scientists check that model. This study gives a mathematical result, not a new sighting through a telescope.
Min Sub Lee, Jeong Han Kim, Kyoungchul Kong+1 more
AstronomerThe authors calculate the global 21 cm brightness temperature and power spectrum in a two-component boosted dark matter (BDM) cosmology. Annihilation of the heavier component energizes the lighter one; self-interactions redistribute that energy, providing pressure support that suppresses small-scale perturbations even for MeV-scale dark matter. The predicted reduction in low-mass halo abundance delays the global absorption feature and shifts the power spectrum’s redshift evolution. SKA1-Low forecasts are most sensitive at intermediate light-component fractions. For σ_self1/m_1 ≃ 0.1 cm² g⁻¹, models with 0.2 ≲ r_1 ≲ 0.7 are distinguishable from fiducial ΛCDM at 95% confidence under the optimistic forecast, not an observational detection. Comparisons with existing SENSEI bounds and projected neutrino-experiment searches highlight complementary tests of BDM.
PublicThis study predicts how a particular dark matter model could change the early Universe’s 21 cm signal. In the model, heavier dark matter particles annihilate and give energy to a lighter component. Interactions spread this energy, making it harder for small dark matter structures to form. The calculations predict that an absorption feature in the signal would occur later and that its spatial variations would evolve differently. An optimistic forecast suggests that the planned SKA1-Low observations could distinguish some versions of this model from the standard cosmological prediction, especially when neither dark matter component is too scarce. These are theoretical predictions and forecasts, not detected signatures. The authors also compare this approach with laboratory limits and possible searches using large neutrino experiments.
Young explorerScientists studied an idea about dark matter, stuff in space we cannot see. In their idea, one kind gives energy to another kind, making it harder for little clumps to grow. Their calculations say this could change a radio signal from long ago. A telescope might someday spot the difference, but this study has not found it yet.
Damiano F. G. Fiorillo, Martina La Rosa, Federico Testagrossa+1 more
AstronomerThe authors reassess constraints on nonstandard neutrino magnetic dipole moments from the absence of a high-energy neutrino signal from SN 1987A. In the proposed mechanism, electromagnetic fluctuations in the supernova plasma convert some neutrinos into sterile right-handed states; these escape with energies around 100 MeV and can reconvert into standard neutrinos in the Galactic magnetic field. Incorporating scattering on both protons and electrons, advances in supernova simulations, and updated descriptions of the Milky Way magnetic field, the analysis derives μνe ≲ 4.8 × 10⁻¹³ μB for Dirac electron neutrinos. The authors report that this inferred bound supersedes existing terrestrial and astrophysical constraints; it is not a detection of a magnetic moment.
PublicNeutrinos might have a tiny magnetic property that lets them change into a sterile form inside an exploding star. Under this scenario, some would escape and change back in the Milky Way’s magnetic field, producing a high-energy neutrino signal. Such a signal was not observed from supernova SN 1987A. The researchers use that absence, together with updated supernova simulations and models of the Galaxy’s magnetic field, to calculate a tighter limit on this possible property. For electron neutrinos of the Dirac type, they infer a magnetic moment no greater than about 4.8 × 10⁻¹³ Bohr magnetons, which they report is the strongest existing limit. This does not establish that neutrinos have such a magnetic moment.
Young explorerAn exploding star sends out tiny particles called neutrinos. Scientists wondered whether a tiny magnetic property could let some of them change into a hidden kind and then change back on their way here. The expected high-energy signal was not seen from one past star explosion. Using computer models, scientists worked out a stricter limit on how strong that possible property could be. They did not discover that the property exists.
Gonzalo Herrera, Filippo Sala, Piero Ullio+17 more
AstronomerThis white paper reviews the prospects for probing physics beyond the Standard Model through multi-messenger and multi-wavelength observations of active galactic nuclei (AGN). Motivated in part by IceCube associations of neutrinos with individual AGN, it surveys AGN physics, surrounding dark matter, and their interface with BSM searches. It identifies two parallel priorities: improving understanding of AGN and their dark-matter environments, and finding promising BSM signatures that remain robust despite uncertainties in that understanding. The paper outlines key open questions rather than reporting a BSM discovery.
PublicActive galactic nuclei are being explored as possible laboratories for discovering physics beyond the Standard Model. This white paper reviews how observations using different kinds of cosmic signals and different wavelengths of light could help, motivated partly by IceCube's associations of neutrinos with individual active galactic nuclei. The authors emphasize that researchers must better understand these sources and the dark matter around them, while identifying possible signs of new physics that would remain convincing despite gaps in that knowledge. This is a review of opportunities and unanswered questions, not a report of a discovery.
Young explorerScientists want to use light and tiny particles coming from places called active galactic nuclei to look for new rules of nature. But first, they need to understand these places better, including the mysterious dark matter around them, so they do not mistake something ordinary for something new. This paper explains what scientists know and what they still need to figure out. It does not say they have found new rules yet.
Jun-Ye Chen, Dongdong Wei, Zong-Kuan Guo
AstronomerThree-dimensional lattice simulations of two coupled scalar fields investigate gravitational-wave production during a cosmological first-order phase transition with fluctuation-induced friction. Kinetic and gradient energy in the additional field s accumulates near expanding phi bubble walls and alters their motion. The s contribution dominates the gravitational-wave spectrum at its peak, with a slope resembling the bulk flow model over a finite interval above the peak. At high wave numbers, phi dominates and the total spectrum develops a broad feature. For the larger coupling studied, slowly contracting false-vacuum remnants survive long after bubble merger. These are simulation findings, not observational detections.
PublicResearchers used three-dimensional computer simulations to study how an early-universe transition could produce gravitational waves. Their model contains two interacting fields, with bubbles of a new state expanding in one field. Energy in the second field gathers near the bubble boundaries and changes how they move. This second field contributes more to the gravitational-wave signal at the peak of its spectrum, while the bubble-forming field dominates at shorter wavelengths, where the combined spectrum develops a broad feature. With the stronger interaction studied, pockets of the old state shrink slowly and remain long after bubbles merge. The findings describe the simulated model, not an observed cosmic event.
Young explorerScientists made a computer model of a change that might happen in the young universe. In the model, bubbles grow and join together. Something else in the model gathers energy around the bubble edges and changes how they move. The scientists studied the ripples in space that this could make. When the two parts of the model interact more strongly, little pockets of the old state stay around for a long time, slowly shrinking. This happened in a computer model, not in something the scientists watched in the sky.
Alice Garoffolo, Gianmassimo Tasinato
AstronomerThis theoretical study develops a scalar-wave framework for corrections to Newtonian point-mass lensing, showing sensitivity to both the exterior radial potential and the inner boundary condition. Corrections are organized in inverse powers of lens distance using dressed radial kernels that incorporate the Newtonian interaction to all orders and resum the long-range phase that invalidates a free-wave Born expansion. With wavefront curvature retained, an inverse-square deformation changes the direct wave by an amount equal to the scattered-wave contribution, even when scattering is faint. Steeper potential terms produce short-distance divergences requiring local terms fixed by the inner boundary condition. For a Schwarzschild black hole, matching to an ingoing horizon solution yields an absorptive contact term that reproduces the horizon-area cross section. This term is absent for a non-absorbing lens of equal mass and lies outside the thin-lens diffraction integral. It completes the mass-quadratic correction, entering at the same order as finite-distance optical corrections and largely cancelling against them for an aligned source. These are analytic results within a scalar-wave treatment, not observational findings.
PublicA theoretical study asks how gravitational waves passing a massive object could be affected by more than its total mass. Using a simplified scalar-wave model, the researchers calculate effects from both the object's surrounding gravitational field and the condition imposed at its centre. They find that counting only scattered waves can miss an equally large change in the direct wave for one type of correction. For a non-rotating black hole, their calculation also includes absorption at the horizon, an effect absent for a non-absorbing object of the same mass and not captured by the usual thin-lens calculation. This absorption contribution largely cancels with certain finite-distance corrections when source and lens are aligned. The work provides a theoretical framework, not a reported detection.
Young explorerHeavy things can bend waves that pass nearby. Using maths and a simplified kind of wave, scientists studied how this happens. Their calculations say that the heavy thing's mass is not the whole story: whether it swallows waves matters too. They calculated this for a black hole; they did not report seeing it happen.
Haowen Zhong, Ethan Vislosky, Emanuele Berti
AstronomerStochAXtic computes the dimensionless stochastic gravitational-wave background spectrum Ω_GW(f) from compact binary coalescences without Monte Carlo population sampling. It exploits waveform scale invariance: at fixed mass ratio, spins and tidal deformabilities, total mass enters only through Mf. Expensive LALSuite waveform evaluations are cached in a population-independent lookup table, allowing mass and redshift integrals to be recast as FFT-based convolutions. Truncated singular value decomposition reduces storage and computational costs. After table construction and caching, the authors report evaluations in a few milliseconds, relative numerical errors below 10^-3 at frequencies up to 1 kHz, and agreement with popstock within its statistical uncertainties. The JAX implementation provides derivatives with respect to population hyperparameters for use in gradient-based inference, including Hamiltonian Monte Carlo. These are computational performance and validation results, not a gravitational-wave detection.
PublicThe authors introduce StochAXtic, software that calculates the predicted combined gravitational-wave background from merging pairs of compact objects. Instead of repeatedly sampling possible populations, it reuses a table of precomputed waveforms and applies fast mathematical methods to combine their contributions. The authors report that, once this table is built and saved, a spectrum takes only a few milliseconds to calculate, with relative numerical error below 0.1% at frequencies up to 1,000 hertz. Its calculations agree with another program, popstock, within that program’s statistical uncertainties. StochAXtic also calculates how the predicted background changes when population assumptions change, making it suitable for certain methods of inferring those populations. The work presents a calculation tool, not a new observation of gravitational waves.
Young explorerWhen pairs of very small, heavy space objects join together, they make ripples in space. Many ripples together can make something like a quiet hum. The researchers made a computer tool to work out what that hum might be like. It saves tricky calculations so it can use them again and give answers very quickly. They checked its answers against another tool and found that they matched within how sure that other tool could be. This is a new way to calculate the hum, not a report that they heard it.
Davide Ciattaglia, Adrien Kuntz, Enrico Barausse
AstronomerThis theoretical and numerical study gives a geometric interpretation of convergence limits for quasinormal-mode overtone sums in Kerr black-hole ringdown. Although highly damped overtones decay faster in time, their Green-function amplitudes grow exponentially with the optical distance of the source and observer from a characteristic bounce radius, where the mode amplitude is smallest. For one mode family, amplitudes are additionally suppressed away from the equatorial plane, making convergence depend on direction as well as distance and extending the picture to bounce surfaces. Combining these effects yields a sufficient condition for absolute convergence, not a necessary one. For a/M = 0.7, the authors map the position-dependent convergence time; numerically computed amplitudes reproduce the predicted spatial dependence to better than one percent. These results constrain when overtone sums converge rather than reporting a gravitational-wave observation.
PublicA black hole’s fading gravitational-wave signal is often modeled by adding many mathematical ringing patterns. Adding more patterns does not always improve the description: the faster-fading patterns can have increasingly large amplitudes, so the sum may not converge until enough time has passed. This theoretical study links that timing to the positions of the signal’s source and observer around a rotating black hole. Distance matters, and for one family of patterns, direction relative to the black hole’s equator matters too. The researchers derive a condition that guarantees convergence and test their predicted position dependence with numerical calculations for one black-hole spin, finding agreement to better than one percent. The work establishes limits on the mathematical model, rather than measuring a new black-hole signal.
Young explorerScientists describe a black hole’s fading wiggles by adding lots of smaller wiggle patterns together. But adding more patterns does not always make the math work: sometimes they must wait until later in the fading signal. In this study, scientists used math and computer calculations to show how that waiting time depends on where the wiggles start and where someone watches, around a spinning black hole. This helps them know when their way of describing the wiggles will work.
Linghao Song, Jingtao You, Ming Cheng+2 more
AstronomerThis theoretical study computes chemical-potential-enhanced cosmological collider signatures from right-handed-neutrino loops coupled to the inflaton. In the collinear squeezed configuration, the triangle loop factorizes into hard and soft parts. The authors evaluate the soft part using complete mode functions and the full loop integral, while treating the hard part as a tree-level subdiagram. Subtracting large-momentum asymptotic terms while retaining their finite oscillatory contributions, and performing the full spinor-helicity and Schwinger-Keldysh sums, yields a finite, closed analytic expression for the complex clock coefficient. They also derive a compact leading-signal formula in the large-chemical-potential expansion. A Mellin-pole analysis attributes half the asymptotic coefficient to the first local pole on each soft line along one Schwinger-Keldysh path, and the other half to the remaining local poles on the second soft line. These are theoretical calculations, not observational detections.
PublicThe researchers calculate possible cosmological signals produced by hypothetical right-handed neutrinos interacting with the inflaton. Their focus is an oscillating pattern whose strength can be enhanced by a quantity called chemical potential. They separate a complicated particle-loop calculation into two parts, evaluate one in detail, and treat the other with a simpler mathematical description. This produces a finite analytic formula for the oscillating signal, plus a more compact expression valid in the large-chemical-potential limit. They also identify how different mathematical contributions combine to produce the result. The work predicts signals within a theoretical model; it does not report finding right-handed neutrinos or detecting these patterns.
Young explorerScientists used math to figure out what kind of wiggly pattern some possible tiny particles might make in the very young universe. They split a hard calculation into smaller pieces and worked out how the pieces fit together. They calculated what a pattern could look like; they did not find the particles or see the pattern.
Luis A. Escamilla, Özgür Akarsu, Eleonora Di Valentino+2 more
AstronomerThe authors test whether Lambda_sCDM, which posits a rapid cosmological-constant sign switch from negative to positive near redshift 2, can jointly alleviate the growth-index, H0, and S8 tensions when gamma is free. For PL18+fσ8, gamma-Lambda_sCDM reduces the gamma and H0 discrepancies from 3.7σ and 4.2σ in gamma-LambdaCDM to 2.1σ and 1.1σ, respectively, while maintaining low S8 and essentially unchanged minimum chi-squared; the transition redshift is constrained to exceed 1.58 at 95% confidence. Among free-transition-redshift analyses, PL18+fσ8+PP&SH0ES shows the largest improvement, yielding a transition redshift of 1.82 (+0.32, −0.28) at 95% confidence, residual gamma and H0 tensions of 1.8σ and 0.6σ, and an improved fit. However, SDSS-IV/eBOSS and first-year DESI 3D BAO data favor transition redshifts above roughly 2.3–2.4, weakening the joint alleviation, especially for H0. Fixing the transition at redshift 1.7 gives low residual tensions for PL18 and PL18+fσ8 but somewhat worse fits; free-redshift analyses exclude that benchmark at more than 2σ when uncalibrated PantheonPlus supernovae or 3D BAO data are included. These are dataset-dependent model-fitting results, not a detection of the proposed transition.
PublicDifferent cosmological datasets do not fully agree about how fast the universe expands or how cosmic structure grows. This study tests a proposed model in which the cosmological constant changes rapidly from negative to positive. For some combinations of data, allowing this change reduces disagreements about both expansion and structure growth while keeping another measure of cosmic clustering low. The improvement is not universal: measurements of large-scale cosmic patterns favor a different transition time and weaken the benefit, especially for the expansion rate. A particular fixed transition time also gives somewhat worse fits and is disfavored when certain additional data are included. The results suggest that the proposal may ease several cosmological disagreements, but do not establish that the change actually happened.
Young explorerScientists have different clues about how the universe grows, and some clues do not agree. They tried a new idea in their calculations: what if something that affects the universe’s expansion changed from negative to positive? This helped some clues agree better, but helped less when other clues were added. They have not shown that this change really happened.
Guo Chen, Wei Fang, Chao-Jun Feng
AstronomerThe study analyzes 35 HHDE fluid realizations combining five bulk-viscosity cases with seven dark-sector interactions, with density terms proportional to H² and H⁻². Reconstructing intrinsic dark-energy pressure from the continuity equation makes dark-energy viscosity invisible to the background expansion; interaction and dark-matter viscosity introduce further exact degeneracies through their net contribution. All realizations admit stable de Sitter attractors in physical parameter domains, either dark-energy dominated or retaining dark matter sustained by exchange and viscous stress. Representative posterior trajectories approach these endpoints. Fitting supernovae, cosmic chronometers, DESI BAO and a compressed Planck distance prior, the density-product interaction gives H₀ = 71.15 (+1.53/−1.64) km s⁻¹ Mpc⁻¹. Conditional on that prior being applicable, the preferred dark-energy-to-dark-matter transfer lowers minimum χ² by 75.16 relative to noninteracting HHDE and leaves approximately a one-standard-deviation discrepancy with the local distance ladder when both uncertainties are combined.
PublicThis study tests mathematical models in which dark energy and dark matter can exchange energy and experience a viscosity-like effect. Some different choices produce exactly the same cosmic expansion history, so that history alone cannot distinguish them. The models also allow stable future states with a constant expansion rate, sometimes with dark matter remaining. These are theoretical outcomes, not observations of the future. Combining several astronomical datasets with a compressed summary of Planck distance information, one interaction model favors energy flowing from dark energy to dark matter and estimates today's expansion rate at 71.15 (+1.53/−1.64) kilometres per second per megaparsec. If that Planck summary is applicable to the model, the fit improves over the version without energy exchange, and the expansion-rate estimate is within about one combined standard deviation of the local distance-ladder measurement. This is a conditional model-based inference, not a direct detection of energy transfer.
Young explorerScientists made math stories about how the universe grows. In these stories, two mysterious things called dark energy and dark matter can share energy, and a kind of internal friction can matter too. Different stories can make the universe grow in exactly the same way, so they can be hard to tell apart. The math allows futures where the universe keeps growing at a steady rate. One story matched clues from the sky better when energy moved from dark energy to dark matter, but that answer depends on whether an important piece of information can be used for this story. It does not prove that the energy really moves that way.
Yogesh, Imtiyaz Ahmad Bhat, Mayukh R. Gangopadhyay+1 more
AstronomerThe ACT-inclusive P-ACT-LB dataset combination gives a scalar spectral index ns = 0.9743 ± 0.0034, placing standard quintessential inflation close to or beyond the 2σ boundary in the tensor-to-scalar ratio–spectral index plane. This theoretical study revisits quintessential inflation in Einstein–Gauss–Bonnet gravity, with a scalar field non-minimally coupled to the Gauss–Bonnet invariant. Of three coupling functions examined, exponential and hyperbolic-secant couplings can shift predictions into the 1σ region allowed by P-ACT-LB-BK18; the hyperbolic-tangent coupling remains disfavoured. A model-independent reheating parameterisation also permits viable thermal histories without a potential minimum, with reheating temperatures consistent with Big Bang nucleosynthesis bounds. These are model-viability results, not detections of the proposed coupling. Their quantitative significance depends on dataset choice: ACT and, indirectly, DESI BAO drive the higher ns preference, whereas Planck-only combinations or those including the latest SPT-3G data yield lower central ns values and milder tension for standard quintessential inflation.
PublicSome combinations of observations favour a pattern of early-universe fluctuations that challenges a model called quintessential inflation. The researchers explore whether modifying gravity could make that model fit better. Their calculations test three mathematical forms for how a field interacts with gravity. Two can bring the predictions into close agreement with one combined dataset; the third remains disfavoured. They also find that the model can allow a reheating stage compatible with limits from the formation of light elements in the early universe. This does not establish that the modified theory is correct. The apparent need for it depends on which observations are combined: other combinations put less pressure on the original model.
Young explorerScientists use clues from space to test ideas about how the universe grew very quickly when it was young. These researchers tried three changes to the maths in one idea. Two could make it match one set of clues better, but the third did not fit well. That does not prove the changes are right: other sets of clues make the old idea look less troubled.
Matteo Mescia, Héctor Gil-Marín
AstronomerThe authors constrain a phenomenological dark-radiation framework with free-streaming and fluid-like components using Planck CMB, DESI DR2 BAO, and Pantheon+ and DES Y5 supernova data. CMB alone gives N_fld < 0.66 at 95% confidence; CMB+BAO yields N_fs = 2.93 ± 0.23 and N_fld = 0.36 (+0.16/−0.21) at 68% confidence, with the free-streaming component consistent with Standard Model expectations. Within this framework, enhanced early expansion reduces the sound horizon and renders the CMB+BAO discrepancy with SH0ES statistically non-significant according to the T-statistic. Bayesian support is strongly data-dependent: without SH0ES, dark radiation is weakly disfavoured relative to ΛCDM, whereas including SH0ES produces decisive evidence in its favour. The CMB+BAO+Pantheon++SH0ES fit gives N_tot = 3.63 (+0.13/−0.15), f_fs = 0.392 ± 0.026, r_d = 141.8 (+1.3/−1.2) Mpc, and Y_He = 0.2530 ± 0.0017. The inferred helium fraction is approximately 2–2.5 sigma above direct metal-poor H II-region determinations, though broadly consistent with other abundance measurements within current uncertainties. These are model-dependent constraints and comparisons, not a definitive detection of dark radiation.
PublicCould extra radiation in the early universe help explain why different methods disagree about how fast the universe is expanding today? The authors test a model containing both freely travelling radiation and radiation that behaves like a fluid against observations of ancient cosmic light, galaxy clustering and supernovae. In this model, extra radiation speeds up early expansion and shrinks the cosmic distance scale used in the analysis. This makes the expansion estimate from the ancient light and galaxy clustering statistically compatible with SH0ES measurements under the authors’ chosen test. But the evidence depends on which data are included: without SH0ES, the comparison weakly favours the standard cosmological model; with SH0ES, it decisively favours the dark-radiation model. A combined fit also implies more primordial helium than some direct measurements suggest, although other abundance measurements remain broadly compatible. The study therefore offers a possible explanation, not proof that extra dark radiation exists or that the disagreement is settled.
Young explorerScientists have different ways to work out how fast the universe is growing, but their answers do not quite match. This study tried an idea: maybe the young universe had extra invisible radiation, with some moving freely and some acting like a fluid. In the scientists’ calculations, that extra radiation can help the answers agree. But whether the idea looks better than the usual one depends on which measurements they use. It also suggests more helium, a kind of gas, than some measurements find. So it is an interesting idea to keep checking, not something scientists have proved.
Javier de Cruz Pérez, Darío Jaramillo-Garrido, Antonio L. Maroto+1 more
AstronomerThe authors introduce field-theory models unifying the dark sector as a single perfect fluid with vanishing sound speed. The framework uses one scalar field invariant under transverse diffeomorphisms and naturally implements mimetic-model constraints. It also admits an interpretation as cold dark matter interacting with vacuum energy, avoids phantom behavior, and includes standard ΛCDM as a special case. In one simple example, the authors find effectively evolving dark energy that can reproduce the DESI best-fit CPL evolution with the same number of free parameters. This is a theoretical modeling result, not an observational detection of dark-sector interactions or evidence favoring the framework over ΛCDM.
PublicThe researchers propose a theoretical framework that describes dark matter and dark energy together as one fluid. It can also be interpreted as dark matter exchanging energy with vacuum energy, and it includes the standard cosmological model as a special case. One example allows the effective dark energy to evolve and can reproduce the evolution described by DESI’s best-fit CPL model, using the same number of adjustable parameters. This shows what the proposed model can reproduce; it does not establish that dark matter and dark energy actually interact.
Young explorerScientists made a new math idea for describing two mysterious parts of the universe, called dark matter and dark energy, together. One version can copy a pattern of changing dark energy found in a model fitted to observations. But being able to copy that pattern does not mean the new idea is how the universe really works.
Long-long Feng, Tengpeng Xu, Tian-Cheng Luan+7 more
AstronomerHermes is an in situ multiresolution framework that replaces explicit tuple counting in discrete catalogues with algebraic operations on window-filtered continuous density fields. A compact scaling-function reconstruction supports counts-in-cells and two- and higher-point correlation statistics, with alternative window kernels defining generalised statistics without estimator redesign. Its open-source Python implementation, PyHermes, combines FFT-based convolution, MPI/thread parallelism, and GPU acceleration for high-order field contractions. Supported applications include anisotropic two-point statistics, marked correlations, three-point multipoles, and derived fields constructed with Fourier-space differential operators. Demonstrations and efficiency/scalability benchmarks use halo catalogues from cosmological N-body simulations, rather than observed survey catalogues. The central computational advantage is reuse of one reconstructed field across standard and customised statistics; application to large current and future galaxy-survey data sets is an intended use.
PublicThe researchers introduce Hermes, a computational framework for measuring how matter is clustered across the universe. Instead of repeatedly counting pairs, triples, and larger groups of catalogue entries, it turns discrete positions into a continuous density field and applies mathematical filters to measure different patterns. Its open-source Python software, PyHermes, can reuse that field for many kinds of clustering measurements and uses parallel computing and graphics processors for parts of the calculation. The team demonstrates the method and benchmarks its computing performance using catalogues from universe simulations, not observations. The framework is designed to help analyse the large data sets expected from current and future galaxy surveys.
Young explorerImagine a huge picture made of dots. Scientists made a computer tool that turns the dots into a smooth map, then uses different filters to study patterns without counting every little group of dots again. They tested it with dots from computer-made universes. They hope it will help study real galaxy maps.
Gregory Lukens, Donghui Jeong
AstronomerThe authors apply the total-angular-momentum (TAM) formalism to redshift-space clustering beyond the flat-sky approximation, separating radial and angular contributions while incorporating relativistic effects. They present a new parameterization of wide-angle predictions for comparison with survey data and introduce PowerFull, a modification of the Julia-based 2-FAST package for efficient computation of the full relativistic angular power spectrum. Fisher-matrix forecasts indicate that a SPHEREx-like survey could achieve σ(f_NL) ≈ 1, but omitting relativistic and wide-angle contributions would shift the inferred f_NL by roughly its forecast uncertainty. These are forecasts, not survey measurements, and indicate that consistent modeling is needed to avoid bias at this precision.
PublicThe arrangement of galaxies across enormous distances could reveal whether the Universe’s initial fluctuations departed from a Gaussian pattern, described by a parameter called f_NL. But surveys covering broad areas of sky must account for both relativistic effects and the different directions in which galaxies are viewed. The authors develop a mathematical description that handles these effects and introduce software called PowerFull to calculate the predicted clustering patterns efficiently. Their forecasts suggest that a survey like SPHEREx could measure f_NL with an uncertainty of about 1. Ignoring these effects could shift the answer by about that same amount. This is a prediction of measurement performance, not a detection of a signal from the early Universe.
Young explorerScientists want to use huge maps of galaxies to learn how the Universe began. But when we look in many directions, the patterns need extra-careful math so we do not get the wrong answer. These researchers made a way to do that math and a computer tool to help. Their calculations suggest that leaving out those details could change a future answer by about as much as its expected wiggle room. They have not measured that answer yet.
N. Sanders, H. Seo, M. Rashkovetskyi+65 more
AstronomerThe authors combine overlapping DESI DR2 BAO tracers into a unified catalog using linear-bias weights, avoiding double counting of cosmic volume while incorporating auto- and cross-information at catalog level. Updates include quasars out to z=1.6 and redshift-dependent bias treatment to improve reconstruction. The resulting distance measurements are highly consistent with the baseline DR2 analysis, and tests find no evidence of tracer-dependent systematics over 0.8<z<1.6. Finer redshift binning reveals no indication of a previously missed expansion-history feature in that interval and reproduces DESI’s preference for an evolving dark energy equation of state, rather than establishing such evolution.
PublicThe researchers combined overlapping DESI samples of galaxies and quasars into one catalog to measure cosmic distances without counting the same cosmic volume twice. They extended the method to include additional quasars and adjusted how the objects are weighted across redshift. Their distance measurements closely agree with the standard DESI DR2 analysis. Over the redshift range 0.8 to 1.6, they found no evidence that different types of objects introduce different systematic errors. Examining distances in finer redshift slices showed no sign of a previously missed feature in the universe’s expansion history. The analysis also preserves DESI’s preference for dark energy whose properties change over time, but does not establish that they do.
Young explorerScientists joined lists of galaxies and objects called quasars to make one map, taking care not to count the same part of space twice. Their new distance measurements agreed closely with the earlier ones. Looking in smaller steps, they found no sign of a hidden change in how the universe was growing in the part they checked. The results still hint that dark energy might change over time, but they do not prove it.
Sabir Ramazanov, Zhen-Min Zeng
AstronomerThe authors explore a possible early-Universe scenario in which domain walls temporarily dominate the energy density, interrupting radiation domination before annihilating into particles. During wall domination, expansion accelerates while the network’s comoving correlation length stays constant. Tensor modes sourced by the walls freeze upon horizon exit and re-enter during the subsequent radiation era, contributing to a gravitational-wave background. The predicted spectrum has a distinctive plateau in place of the peak characteristic of sub-dominant walls. The authors argue that the signal could be observable with present-day and future facilities across a broad frequency range. Its energy density could also significantly affect the effective number of neutrino species unless the duration of wall domination is sufficiently constrained. These are theoretical predictions, not reported detections.
PublicThe paper explores a possible episode in the early Universe when structures called domain walls briefly controlled cosmic expansion before disappearing into particles. In this proposed scenario, expansion speeds up and the walls generate gravitational waves, or ripples in spacetime. The predicted gravitational-wave spectrum contains a flat region instead of the peak expected when the walls never dominate. The authors argue that these waves could be strong enough for current or future facilities to observe over a wide range of frequencies. The waves could also carry enough energy to significantly change the effective number of neutrino species, a measure used to describe the Universe’s radiation content, unless the wall-dominated episode was limited in duration. The abstract reports a theoretical possibility, not an observation of such an episode.
Young explorerScientists are exploring an idea about when the Universe was very young. Maybe it had things called domain walls that briefly made its growth speed up, then disappeared into tiny particles. In this idea, the walls also made ripples in space. Scientists predict those ripples would have a special pattern that we might be able to find. If the walls stayed in charge too long, the ripples could carry a lot of energy and change how scientists describe the young Universe. This is an idea to test, not something they have seen happen.
A. Coerver, M. Rahimi, D. Anbajagane+92 more
AstronomerUsing two years of SPT-3G data, the authors search for anisotropic cosmic birefringence: a hypothesized, direction-dependent rotation of CMB polarization that mixes E and B modes. They measure the rotation-angle power spectrum in six bins over 20 < L < 2048 and find it consistent with zero, with no evidence for birefringence. For a scale-invariant spectrum, they report a new best 95% confidence upper limit of A_CB < 0.027, and also constrain a scale-dependent white-noise model. Translating the scale-invariant limit into physical constraints gives 95% confidence bounds of g_{αγ} < 1.7 × 10^-2/H_I and B_{1 Mpc} < 9.6 nG. These translated bounds depend on the corresponding physical interpretation. Several-fold improvement with the complete SPT-3G survey is forecast, not demonstrated.
PublicResearchers used two years of observations from the South Pole Telescope’s SPT-3G instrument to test whether light from the cosmic microwave background has had its polarization—the orientation of its oscillations—twisted by different amounts in different directions across the sky. They found no evidence for this proposed effect, called anisotropic cosmic birefringence. The result does not prove the effect is absent, but places tighter limits on how strong it could be under a particular model of its variation across the sky. The team also translates that limit into bounds on axion coupling and primordial magnetic-field strength under the corresponding physical interpretations. They expect the full SPT-3G survey to improve the limits several-fold; that improvement has not yet been established.
Young explorerScientists used a telescope at the South Pole to look at very old light in the sky. They wanted to see whether something on its long journey had twisted the way the light wiggles, with different twists in different parts of the sky. They did not find a clear sign of that twist. A twist might still be too small to spot, so their work helps show how small it would need to be. They hope more observations will let them check even more carefully.
Sara Bianco, Paul Frederik Depta, Jonas Frerick+3 more
AstronomerThis theoretical study derives cosmological constraints on a relic particle decaying into a neutrino–antineutrino pair, focusing on late-time destruction of light nuclei synthesized during big-bang nucleosynthesis. Final-state radiation and subsequent interactions of injected neutrinos with thermal-background neutrinos or one another can generate electromagnetic and often hadronic secondaries, inducing photodisintegration and hadrodisintegration. A Monte-Carlo-inspired probabilistic treatment, rather than Boltzmann techniques, accounts for these processes and their interplay. The authors report significant constraints from observed light-element abundances across a broad range of previously unexplored relic masses and lifetimes; this is not a detection of the proposed particle.
PublicThe observed amounts of light elements in the early Universe can help test whether a hypothetical leftover particle could have decayed into neutrinos. The researchers calculate how radiation accompanying the decay, together with later neutrino interactions, could produce material capable of breaking apart light atomic nuclei that had already formed. Using a probabilistic calculation to track these linked processes, they find strong limits on possible combinations of the particle’s mass and lifetime, including a broad range not previously explored. The study constrains a hypothetical scenario rather than finding evidence that the particle exists.
Young explorerLong ago, the Universe made tiny centers of atoms. Scientists wondered whether a mystery particle could have broken into other tiny particles and started a chain of events that smashed some of those centers. They used calculations and the amounts of light elements we see to work out which versions of this idea would cause too much smashing. They did not find the mystery particle.
Miguel Escudero, Thomas Hambye, Chandan Hati
AstronomerThe authors propose an asymmetric dark matter scenario with two new fields, a scalar singlet and an inert scalar doublet. Renormalizable interactions slowly transfer an asymmetry from the Standard Model Higgs sector to dark matter. Successful transfer requires small, technically natural couplings, making the model effectively CP conserving. The proposed model permits direct detection in next-generation experiments and predicts an inert-doublet mass of 46–580 GeV, potentially accessible at the LHC and future colliders. Direct-detection measurements may also allow its mass and lifetime to be inferred, pointing to long-lived, compressed electroweak states searchable at the HL-LHC. These are model predictions and prospective tests, not reported detections.
PublicThe paper proposes a relatively simple explanation for how dark matter could have developed an excess of particles over antiparticles. It adds two new particle fields and uses very weak interactions to gradually pass an imbalance associated with the Higgs field to dark matter, without requiring a new source of the particle–antiparticle symmetry violation known as CP violation. The model could be tested by the next generation of experiments searching directly for dark matter and by particle colliders. It predicts a new particle sector with masses between 46 and 580 GeV. Future direct-detection results might also reveal the mass and lifetime of the proposed collider particles, helping guide searches at the upgraded LHC. No discovery is reported.
Young explorerScientists have a new idea about dark matter, the invisible stuff they want to understand. Their idea adds two new kinds of particles and suggests how tiny interactions could leave more dark matter particles than their opposite partners. If the idea is right, future experiments might find dark matter or make some of the new particles. The scientists have proposed a way to look, but they have not found these particles yet.
Matias P. Gonzalez
AstronomerThis theoretical work formulates WIMP freeze-out using a dynamical Tsallis parameter qχ(x), with x=mχ/T, allowing relaxation toward extensivity to remain incomplete at chemical decoupling. During freeze-out, the authors adopt an extensive Standard Model bath and specify statistical weights and an effective kinetic closure for the dark matter abundance. Derivatives of the equilibrium density include the evolution of qχ, modifying how rapidly the abundance must change to track its instantaneous reference state. The residual freeze-out value then feeds a phenomenological memory channel into the electron-positron and neutrino sectors, while photons remain extensive. The authors derive the associated chain-rule contributions to radiation energy and entropy balances and formulate photon reheating without automatically imposing conservation of comoving electromagnetic entropy when q varies. This construction links freeze-out history to the neutrino-to-photon temperature ratio and N_eff, subject to the stated sectorial and closure assumptions. No numerical results or observational fits are presented.
PublicThe authors propose a mathematical framework for how a possible type of dark matter stopped maintaining its equilibrium abundance as the early universe cooled. Their model lets the dark matter’s statistical behavior gradually approach the standard form, without requiring that transition to finish before freeze-out. They then introduce a proposed way for a remnant of that behavior to affect electrons, positrons and neutrinos, while photons retain standard statistics. The framework connects this history to the relative temperatures of neutrinos and photons and to a quantity called N_eff that describes the universe’s radiation content. The treatment carefully accounts for changing statistics when tracking energy and entropy. These are theoretical connections based on specified assumptions, not measured effects: the paper provides neither numerical results nor comparisons with observations.
Young explorerImagine tiny dark matter particles in the very young universe following rules that slowly change. Scientists wrote a math model in which traces of those earlier rules might later affect how warm some other tiny particles are compared with light. This is an idea to explore, not something they showed really happened.
Giordano Cintia, Federico Piazza, Samuel Ramos
AstronomerThis theoretical calculation derives the commutator of a scalar field minimally coupled to gravity at leading order in Newton’s constant G_N. The operator-valued corrections contain derivatives of Dirac delta functions supported on the Minkowski light cone. Evaluating these terms on classical/coherent graviton states recovers the causal structure of classical curved spacetime, while their variance introduces uncertainty in causal relations. For a thermal graviton state at temperature T, the late-time light-cone displacement variance is Var(δx) = 4G_N Tt/3. This result subtracts a universal, logarithmically ultraviolet-divergent vacuum contribution that is subleading at late times; it is not an observational measurement.
PublicA theoretical study asks how quantum gravity could make the boundary between events that can and cannot influence one another uncertain. The calculation reproduces the usual causal structure of curved spacetime when gravity is treated classically, but also identifies quantum uncertainty in that structure. For a modeled thermal population of gravitons, the predicted variance in the boundary’s position grows with temperature and elapsed time at late times. The result involves subtracting a divergent vacuum contribution that becomes less important at late times. No observational detection is reported.
Young explorerLight sets a limit on how quickly one event can affect another. Using math, researchers studied whether tiny quantum ripples in gravity could make that limit’s position a little fuzzy. In their model, the fuzziness grows over time when the ripples are in a warm state. This is a calculation, not something they have seen happen.
Özgür Akarsu, Leandros Perivolaropoulos, A. Emrah Yükselci+1 more
AstronomerThe authors present Ph-ΛsCDM, a theoretical realization within general relativity of sign-switching dark energy. A phantom scalar on a bounded hyperbolic-tangent potential drives a smooth, late-time mirror AdS-to-dS transition. Its wrong-sign kinetic term, ordinarily regarded as pathological, enables evolution from a negative- to a positive-energy vacuum-like regime; the field can become repulsive before its energy density turns positive. Within the construction, total energy remains positive, the dynamics remain infrared, and the late-time attractor is de Sitter rather than a Big Rip. Addressing multiple cosmological tensions is a proposed potential application, not an observationally demonstrated result.
PublicThe paper proposes a model in which dark energy changes smoothly from negative to positive energy density late in cosmic history. The mechanism uses a hypothetical field with an unusual kinetic-energy sign, a feature normally considered problematic. In this model, the field can have a repulsive gravitational effect even while its energy density is negative. The authors report that the model keeps the universe’s total energy positive and approaches a de Sitter state rather than ending in a Big Rip. They suggest it could help address disagreements in cosmology, but the abstract does not establish that it resolves them or that observations support the model.
Young explorerScientists made a mathematical model of how the universe might work. In it, something called dark energy changes from a negative amount to a positive amount. Surprisingly, it can start pushing things apart before that change is finished. In their model, the universe does not end by being ripped apart. The idea uses a strange kind of field that scientists usually worry about, and it has not been shown here to describe the real universe.
Steffen Gielen, Rita B. Neves
AstronomerThis theoretical study examines inflation in unimodular loop quantum cosmology, using a geometrically defined unimodular time rather than a free, massless scalar field as a clock. This permits natural Schrödinger evolution independently of the matter content, with an inflaton included without serving as the clock. Working primarily with semiclassical effective equations, the authors derive analytical solutions for simple cases, including a constant potential, and discuss numerical solutions for quadratic and Starobinsky potentials under different initial conditions. They show that an α-attractor potential admits observationally compatible models with either kinetic- or potential-energy-dominated bounces. Possible observable imprints of the quantum-gravity regime remain a prospect, not a reported detection. Connections to full unimodular loop quantum gravity are also discussed.
PublicThe researchers explore a theoretical description of the early universe that combines inflation, a period of rapid expansion, with loop quantum cosmology. Their approach uses a time coordinate defined by geometry, rather than relying on a matter field to act as a clock. Using approximate equations, they obtain mathematical solutions for simple models and numerical solutions for more complex inflation models, comparing different starting conditions. For one family of models, called α-attractors, they find that a cosmic bounce can be dominated by either kinetic or potential energy while remaining compatible with observations. These models might also leave observable traces of quantum gravity, but the study does not report detecting such traces.
Young explorerThe scientists used math to explore how the very young universe might have grown very fast. In their models, they use a clock based on the universe’s geometry instead of making a kind of matter keep time. Some models let the universe shrink, bounce, and then grow. These ideas can fit what we observe, and the bounce might leave clues we could look for. But the scientists have not found those clues or shown that a bounce really happened.
Shuntaro Aoki
AstronomerThis theoretical study examines how a continuous mass spectrum modifies cosmological collider signatures in primordial correlators. The authors develop a spectral representation in which the soft-limit, non-analytic seed function is a Fourier–Laplace transform of the spectral weight in logarithmic momentum variables. Superposing different clock frequencies can dephase the oscillation characteristic of a single massive exchange. They implement this mechanism in an RS2-like inflationary braneworld with a brane-localized inflaton in a five-dimensional AdS bulk. Its tensor sector includes a localized massless graviton and a Kaluza–Klein graviton continuum beginning at m = 3H/2. Using the derived brane spectral weight in the inflaton four-point function yields a smooth seed function rather than a persistent logarithmic oscillation: the continuum begins at zero clock frequency, and its spectral weight vanishes near threshold. This is a model-based example, not an observational detection, of how continuum spectra can alter cosmological collider signals.
PublicParticles present during the universe’s early inflationary era could leave distinctive patterns in primordial correlations. A single massive particle is usually expected to produce a repeating pattern, but this theoretical study asks what happens when the possible particle masses form a continuous range. The calculations show that the different oscillation frequencies can fall out of step, washing out the sharp repetition. The authors demonstrate this in a model with an extra spatial dimension, where a continuous range of gravitational states contributes to a correlation involving four inflaton fluctuations. In that model, they obtain a smooth mathematical signal instead of a lasting oscillation. This is not an observed feature of the universe; it illustrates how departures from the usual pattern could carry information about a continuous spectrum of high-energy states.
Young explorerImagine many clocks ticking at different speeds. Their ticks can get out of step, so you no longer hear one clear beat. Scientists used math to study a bit like this in a model of the very early universe. In their model, many kinds of tiny particles together make a smooth pattern instead of a repeating wiggle. They have not seen this in the sky; it is something their model shows could happen.
Mateusz Kulejewski, Bogumila Swiezewska, Jorinde van de Vis
AstronomerThis theoretical study examines particle production following strongly supercooled cosmological phase transitions, in which the scalar field typically transitions to a value well below its potential minimum. During subsequent evolution, temporarily tachyonic modes can be efficiently produced, potentially channeling a significant fraction of the released energy into particles rather than field gradients. The authors identify consequences for the shape and amplitude of the gravitational-wave spectrum generated by interactions in bubbles, as well as a new high-frequency contribution. These are theoretical findings, not reported observations of particles or gravitational waves.
PublicA strongly supercooled transition in the early universe could release energy and generate gravitational waves—ripples in spacetime. This study explores how the evolving field responsible for such a transition could turn a substantial share of that energy into particles, rather than energy stored in spatial variations of the field. The process would affect the strength and pattern of the gravitational waves associated with interactions in bubbles and produce an additional high-frequency signal. The abstract describes a theoretical mechanism, not a detection.
Young explorerScientists studied what might happen when the very young universe changed, a little like water changing into ice. Some energy could turn into tiny particles and change the ripples made in space, adding extra fast wiggles. This is an idea scientists studied, not something they say they have seen.