
ALT Fast radio bursts (FRBs) probe the integrated free-electron density along extragalactic sightlines and can be used to study late-time cosmic evolution. Conventional Bayesian analyses require repeated evaluations of computationally expensive likelihoods, making inference increasingly challenging as FRB catalogs grow. In this work, we develop an interpretable attention-based neural posterior estimator (AttNPE) for rapid posterior inference of the physical parameters from FRB catalogs with measured redshifts. The attention encoder compresses the catalog into a low-dimensional representation, and a conditional masked autoregressive flow maps samples from a base distribution to the seven-dimensional posterior. AttNPE yields posterior distributions in close agreement with Markov chain Monte Carlo (MCMC), recovering the parameter correlations in both simulated and observed FRB catalogs. For the observed catalog, AttNPE generates 5Γ10β΄ posterior samples in 0.036 s after training, compared with

ALT The formation and evolution of supermassive black holes (SMBHs) remain major open questions in astrophysics. Astrophysical models of SMBHs invoke black-hole (BH) seeds formed at high redshift, which subsequently grow through gas accretion and mergers. Although observations of high-redshift quasars and active galactic nuclei constrain these scenarios, the mechanisms governing SMBH formation and growth remain uncertain. Future gravitational-wave (GW) observations across the millihertz, decihertz, and audio-frequency bands will provide a complementary and direct probe of SMBH assembly by tracing their merger history. We investigate the extent to which multiband GW observations can reconstruct the mass-redshift distribution of these mergers. We simulate populations of BH mergers predicted by different astrophysical models and identify those observable by the millihertz detector LISA, the decihertz detector LGWA, and next-generation ground-based detectors operating in the audio-frequency ba

ALT A phantom-divide transition favored by recent cosmological observations poses an inverse problem: which field potentials realize the inferred expansion history, and what features are uniquely fixed by it? We reconstruct local potentials for a canonical-phantom scalar pair with standard matter and radiation, targeting Metastable Emergent Dark Energy histories constrained by CMB, DESI BAO, and Type Ia supernova data. The coupled-field Friedmann evolution reproduces the target background to numerical accuracy across the transition, with regular perturbations through the crossing. At the phantom divide, the transition slope sets a dimensionless potential-energy release rate insensitive to the partition of kinetic energy. We identify local potential deformations that preserve both background expansion and linear dynamics. This construction gives a classical field realization of the inferred transition and disentangles observationally fixed energy balances from interaction terms requiring ad

ALT Gravitational waves (GWs) propagate through the universe from sources at high redshift, experiencing the gravitational fields of the intervening matter distribution. Halos composed of ultralight dark matter (ULDM) contain strong density fluctuations, reflecting the interference pattern in the field distribution of the ultralight bosons. In the wave-optics regime, ULDM fluctuations act as a diffractive grating, imprinting frequency-dependent modulations of GWs, in addition to magnification and other gravitational lensing effects. Since lensing depends on the projected surface density, we derive its approximate probability distribution and show how it departs from a Gaussian distribution depending on the halo profile and coherence length. We then quantify how the GW diffraction signatures produced by the granules depend on macroimage magnification and parity. The ULDM interference network imprints stochastic diffraction fluctuations at the O(1)% level on GWs moderately lensed by the host

ALT Pulsar timing arrays (PTAs) have found evidence for a gravitational wave background at nHz frequencies. Inspiralling supermassive black hole binaries (SMBHBs) likely contribute to at least part of the signal, but there may be additional cosmological sources from new physics in the early universe. Disentangling the two is complicated by the prohibitively slow modelling of SMBHB spectra based on realistic population models. Here we introduce binaryfog, a surrogate of up-to-date population models also used by the NANOGrav collaboration, that returns the statistical distribution of bent spectra O(10βΆ) faster than direct computation. We use binaryfog to profile over the six astrophysical parameters in fits of four extensively discussed new-physics models: a dark sector phase transition, cosmic superstrings, inflationary gravitational waves, and domain walls. Compared to the widely adopted power-law description of the SMBHB background, the preference for a phase transition drops from 2.8Ο to

ALT We identify a universal relation between the onset of cosmic acceleration (zβ) and the turnover (zββ) of the linear growth observable fΟβ. In a spatially flat ΞCDM Universe, these two redshifts obey R β‘ (1+zβ)/(1+zββ) = 1.1596, independent of the present matter fraction and the overall expansion and growth normalisations. This provides a new consistency test for ΞCDM, linking both background and growth dynamics. We further establish the conditions under which this extends to broader classes of dark energy and modified gravity theories.

ALT Deviations from general relativity on cosmological scales have been proposed as possible explanations for dark energy and dark matter. Galaxy cluster clustering offers an independent probe of gravitational growth, benefiting from theoretically motivated halo-bias constraints and complementing cluster-abundance measurements. We analyse a subsample of the largest publicly available spectroscopic galaxy cluster catalogue to date, constructed from the DESI Legacy Imaging Surveys. Following our selection cuts, the sample comprises 85 818 clusters at 0.1β€ z β€0.4 with M500cβ³ 6Γ10ΒΉΒ³ Mβ. To enable a robust clustering analysis of this sample, we develop and validate methods for constructing random catalogues and calibrating the effective halo bias. We model the redshift-space two-point correlation function monopole and quadrupole over separations of 10β150 hβ»ΒΉMpc to jointly constrain gravitational growth and AlcockβPaczynski (AP) distortions, fixing the matter power-spectrum shape to that of a f

ALT We describe and test a new numerical method to solve the Vlasov/Boltzmann equation in systems with arbitrary levels of collisionality. The method combines recent ideas to evolve the distribution function (DF) directly in 6+1 dimensional phase-space in a manner which is Lagrangian in both position+velocity. The 'particles' of traditional Monte-Carlo/Particle-In-Cell/N-body (MC/PIC/NB) methods become mesh-generating points defining a volume partition, each of which carries a set of momentum-Lagrangian basis functions representing a local anisotropic momentum DF. The DF is then integrated between cells and bases via a higher-order reconstruction and Godunov-type method, evolving a 10 or 13-moment representation of each basis to ensure manifest mass/momentum/energy conservation. The method allows one to represent arbitrary DFs, including multiple independent streams but also collisional or phase-mixed or thermal distributions (unlike traditional 'moments' or waterbag/tessellation methods),

ALT We update the constraints on Lorentz-violating electrodynamics from the polarization of the cosmic microwave background (CMB). Within the minimal Standard Model Extension, we consider the renormalizable CPT-odd and CPT-even operators of the photon sector, which source cosmic birefringence and circular polarization, respectively. Combining Planck, Atacama Cosmology Telescope (ACT) and South Pole Telescope (SPT) temperature and E-mode data with B-mode measurements from BICEP/Keck, SPT-3G D1 and SPTpol, we constrain the CPT-odd coefficients to |kβ½Β³βΎ(V00)|<7.75Γ10β»β΄β΅ GeV and |k_AF|<3.82Γ10β»β΄β΅ GeV at 95% credible level. These are the strongest bounds to date from any probe, and are driven by the SPT-3G D1 B-mode data. For the first time in this type of analysis, we also include the Planck PR4 EB spectrum, which yields a mild preference for a non-zero |kβ½Β³βΎ(V00)|. For the CPT-even coefficient, we find kF,EβB<5.99Γ10β»Β³Β². Including the latest circular polarization measurements from the Cosmolo

ALT We investigate the thermodynamic evolution of the early Universe across the QCD crossover and up to the electroweak (EW) scale, emphasizing the systematic impact of the QCD equation of state (EoS) and electroweak degrees of freedom on thermodynamic variables, cosmological parameters, and gravitational-wave observables. Using the HotQCD EoS as a baseline, we employ Gaussian Process (GP) regression to generate ensembles of 1000 thermodynamically consistent EoS samples, each extrapolated to high temperatures with and without EW contributions. For each ensemble, we solve the Friedmann equations to obtain the expansion history and compute gravitational-wave (GW) observables, including the instantaneous spectral tilt n_gw(f) and the normalized GW energy density Ξ©_gw(f)/Ξ©_gwΚ³α΅α΅. We find that including the EW sector is the dominant effect across all observables, outweighing differences among QCD baseline models that share the same lattice calibration. In particular, EW contributions drive the

ALT Over the past decade, cosmological observations have started to challenge the ΞCDM model. If these tensions persist, they immediately raise one fundamental question: what new physics could be responsible for them? More specifically, do these deviations arise from a modification of gravity, from an evolving or clustering dark energy component, or from dark matter properties beyond the cold dark matter paradigm? In this paper we show that, with current methods, any observed deviation can always be attributed to more than one sector, preventing us from determining if the tension is due to gravity, dark energy or dark matter. The main reason is that current tests rely on unobservable quantities. These quantities are then either related to observables by assuming the validity of specific equations, or compared with theoretical predictions computed within a given class of theories. Both approaches inevitably break the model independence of the tests, leading to degeneracies between gravity,

ALT The unresolved gamma-ray background (UGRB) provides a promising avenue for probing the cosmological annihilation of weakly interacting massive particle (WIMP) dark matter through its cross-correlation with large-scale structure (LSS). While most previous studies have assumed a velocity-independent (s-wave) annihilation cross section, many well-motivated particle physics models predict annihilation rates that depend on the relative velocity of dark matter particles. In this work, we develop a generalized halo-model framework to investigate the cosmological signatures of velocity-dependent WIMP annihilation. Parameterizing the velocity-weighted annihilation cross section as Ο vα΅£βββ vα΅£βββΏ, we consider four representative cases corresponding to Sommerfeld-like (n=-1), s-wave (n=0), p-wave (n=2), and d-wave (n=4) annihilation. We show that velocity dependence modifies not only the overall annihilation emission but also the halo population dominating the signal, leading to distinct effective

ALT In 2019 the H0LiCOW/TDCOSMO collaboration reported a decreasing Hubble constant Hβ trend with lens redshift in the flat ΞCDM model at low statistical significance (< 2 Ο). The signal is expected if physics is missing from the ΞCDM model in the late universe. In the wake of recent TDCOSMO 2026 results, we revisit the feature. Although the statistical significance of the trend has increased in a like-for-like comparison with earlier studies, propagating mass-sheet transformation (MST) uncertainties reduces the significance to βΌ 1 Ο. As a result, Akaike Information Criterion (AIC) currently favours a constant Hβ over a decreasing Hβ in the ratio 2.4:1. Nevertheless, we highlight good agreement on a negative slope with independent observables and forecast that approximately 40 and 90 lenses will be required to substantiate the trend at βΌ 2 Ο and βΌ 3 Ο significance, respectively, if the feature is physical and current uncertainties are representative. Even now, it is evident that the TDCOSM

ALT Imaging and spectroscopy constrain complementary information about the same sky but are usually processed separately. We introduce Spectro-Spatial Survey Fusion, a framework that unifies these measurements by fitting a continuous spectral surface-brightness field jointly through each instrument's response. Adapting from Gaussian splatting in computer vision, we represent the field as a sum of adaptive spatial Gaussians, each carrying a spectrum built from a shared, learned dictionary of templates. We demonstrate this on 400 Sloan Digital Sky Survey (SDSS) ugriz exposures and 800 fiber spectra across an 84β²Γ84β² region of COSMOS. The joint fit closely reproduces the spatial structure and broad spectral diversity of galaxies, stars, and quasars. For fitted SDSS observations, the summed predicted flux in source-selected pixels agrees with the data to within 2 percent in every band, and 96 percent of fitted spectra agree to within 10 percent in integrated flux. Sparse fiber spectra constrai

ALT Cross-correlating intensity maps with the lensing of the cosmic microwave background (CMB) rejects the foregrounds and noise that do not trace the dark matter. In the near infrared, the remaining signal combines the integrated galaxy light (IGL), the intrahalo light (IHL) and several emission lines. Accurate modeling of all three is essential to forecast detectability and interpret measurements. We present simulated SPHEREx intensity maps of all three on a lightcone, painted on halos from the Hidden Valley N-body simulation, and the CMB-lensing convergence of the same lightcone, over 100 degΒ² to z β 6. We assign each halo its lines and continuum by stellar-population synthesis, in two models whose metallicities are fixed or follow its stellar mass. Halo-model predictions from the simulation's luminosity-halo mass relations and measured clustering, without free parameters, match the auto-spectra of the IGL and lines, and their cross-spectra with lensing, across the band: at 2.2 ΞΌm, to w

ALT The near-infrared extragalactic background light (EBL), which SPHEREx maps in 102 spectral channels, holds the emission lines of galaxies on top of a much brighter continuum from the integrated galaxy light (IGL) and the intrahalo light (IHL). Cross-correlating these maps with the lensing of the cosmic microwave background (CMB) isolates the emission that traces the matter, unbiased by uncorrelated foregrounds. The lines face two obstacles. Filtering along the line of sight removes the long radial modes that carry the cross-correlation, and the extragalactic continuum, a signal in its own right, is a correlated foreground for the lines. We forecast the detectability of all EBL components, in a common halo-model framework, for the SPHEREx deep fields and all-sky survey with lensing maps at the sensitivity level of Planck and the Simons Observatory (SO). Without filtering, and with all amplitudes fitted jointly, the cross-correlation with Planck lensing detects the IGL at 9Ο in the deep

ALT The 21-cm signal from the epoch of reionization is highly non-Gaussian. Standard analyses rely on the power spectrum, which is blind to the Fourier phases, while higher-order statistics such as the bispectrum are computationally expensive and spread information across triangle configurations. As an efficient alternative, we develop the phase gradient spectrum, which measures the circular coherence of phase differences between Fourier modes separated by a long-wavelength mode. We show analytically that this statistic measures the angle-averaged normalised squeezed bispectrum, which describes the response of small-scale power to long-wavelength fluctuations. Applied to reionization simulations with different ionizing source properties, the phase gradient spectrum of the 21-cm signal shows a characteristic non-monotonic evolution. At early times, the spectrum tracks the non-Gaussianity of the matter density field. Once the mean neutral fraction γx_HIγ β² 0.9, the spectrum follows the ioniz

ALT Baryonic feedback is a major theoretical uncertainty for Stage-IV weak lensing (WL) surveys. The thermal Sunyaev-Zel'dovich (tSZ) effect traces the hot gas redistributed by feedback, so combining it with WL can calibrate feedback and improve cosmological constraints. Doing so with simulation-based inference requires correlated lensing and Compton-y maps across many cosmologies and feedback scenarios. We present two methods to generate Compton-y maps from dark-matter-only N-body light cones using the component-wise baryonification (BFC) model, which produces separate gas and stellar fields and computes the gas pressure from hydrostatic equilibrium. The two methods, painting pressure profiles onto halos or redistributing the signal along the baryonified gas, agree to within a few per cent. We calibrate the model to the fiducial and strongest AGN feedback FLAMINGO simulations by jointly fitting gas mass fractions and thermal pressure profiles, measuring the non-thermal pressure from the s

ALT We reconstruct the large-scale radial velocity field from the kinetic Sunyaev-Zel'dovich effect for the first time with the South Pole Telescope, and measure the galaxy-velocity correlation and the velocity power spectrum. We cross-correlate the SPT-3G D1 maps at 95, 150, and 220 GHz with luminous red galaxies from the DESI Legacy Imaging Surveys over the SPT-3G Main field. For the extended galaxy sample, we detect the galaxy-velocity correlation at signal-to-noise ratio 11.2 and the velocity power spectrum at 8.1, combining the three frequencies. We detect the reconstructed velocities for the first time at 220 GHz with signal-to-noise ratios of 4.2 and 3.1 from the galaxy-velocity correlation dipole and monopole, and at 3.8 from the velocity power spectrum. We model and estimate foreground contributions to the monopole, detecting both the thermal Sunyaev-Zel'dovich effect and the cosmic infrared background contribution separately. On the around five percent of sky jointly covered by S

ALT On sufficiently large scales the galaxy overdensity field is approximately Gaussian, so its information can be losslessly compressed into the power spectrum. Standard power spectrum estimators are nevertheless suboptimal on the largest scales, where the survey window induces significant coupling between different modes. Optimal power spectrum analysis typically requires an inverse of the covariance matrix, which can be overly expensive. Here we develop an alternative field level approach, which is significantly faster. We assess its impact on constraining primordial non-Gaussianity of the local type, which is particularly sensitive to these largest scales. Relative to the standard power spectrum analysis, our field level likelihood forecasts a 14% improvement in constraining power. For the DESI DR2 LRG and QSO samples, our approach forecasts Ο(fββ)=4.1, compared to Ο(fββ)=4.8 from the power spectrum based analysis. The field level approach also enables efficient implementations of addi
