KICC papers

arXiv:2607.15020v1 Announce Type: new Abstract: We present new constraints on models of dark matter interacting with neutrino, based on high-resolution Lyman-$\alpha$ forest data. We perform a suite of full hydrodynamical simulations of these models, spanning a range of interaction strengths and thermal histories. We train an emulator on the simulation results. A Monte Carlo Markov Chain analysis yields an upper limit on the interaction strength of $u_{\nu\chi} \leq1.5\times10^{-8}$ (95% C.L.), which is the strongest direct bound to date on such interactions. Our results exclude previous hints of non-zero interactions presented in the literature. We furthermore compare our results to those obtained by mapping warm dark matter constraints to other models with suppressed small-scale structure, and find that these methods would overestimate the constraining power for this model.

arXiv:2603.16991v2 Announce Type: replace Abstract: Galactic haloes host a complex, multiphase circumgalactic medium (CGM), and at high redshift are fed by cold, filamentary inflows. In contrast, mature galaxy clusters are dominated by a hot, enriched, X-ray emitting intracluster medium (ICM), with cold gas largely confined to member galaxies. However, the transition between these regimes remains poorly constrained. We present a cosmological zoom-in simulation of a massive cluster progenitor evolved to $z=2.7$, with enhanced CGM resolution to better trace the accretion, mergers and feedback events that precede the birth of the ICM. We connect this evolution to mock MgII, CII, OVI and OVII absorption, tracing low and high ionisation gas phases. We also study Lyman-$\alpha$ (Ly$\alpha$) and Balmer-$\alpha$ (H$\alpha$) haloes in emission, using radiative transfer in post-processing. Between $z\sim4.4$ and $2.7$, a major merger and AGN feedback drive an inside-out transformation, redistributing gas to larger radii and flattening density, temperature and metallicity profiles. Intermediate column MgII absorbers are rapidly destroyed, leaving a clumpier cold gas distribution associated with satellites, while gas is ionised beyond OVII as the inner halo enters the X-ray regime. An extended Ly$\alpha$ halo remains detectable even without AGN photoionisation, and evolves from filamentary to more spherical as inflowing gas is disrupted. Our fiducial model underpredicts observed central Ly$\alpha$ emission - we likely require more efficient Ly$\alpha$ production in the nuclear region, either through more effective escape of stellar Ly$\alpha$ photons or through enhanced conversion of AGN-powered ionisation into Ly$\alpha$ emission. H$\alpha$ haloes are dimmer and smaller than Ly$\alpha$, but with JWST may provide a complementary probe of the evolving CGM at this critical epoch.

arXiv:2607.14254v1 Announce Type: new Abstract: We present an updated Hubble diagram of giant extragalactic H II regions and H II galaxies spanning the redshift interval $z\sim 0$ to $z\sim 14$, extending the use of the $L(\mathrm{H}\beta)-\sigma$ relation as an independent cosmological probe into the epoch of cosmic dawn. Our sample comprises 243 objects, including local anchor systems with primary distance measurements, previously published low- and intermediate-redshift H II galaxies, recent JWST/NIRSpec observations, and a new subsample of extremely high-redshift H II galaxies observed with JWST and with ALMA+JWST for the highest-redshift cases. We homogenise the measurements across the full sample, applying consistent corrections for instrumental and thermal broadening, extinction, and, where required, the transformation from [O III]-based to Balmer-line velocity dispersions. We model lensing effects of high redshift using a log-normal magnification formalism and infer cosmological parameters via a nested-sampling analysis. The $L-\sigma$ relation remains consistent over the full redshift range, showing no evidence for significant evolution even at the highest redshifts currently accessible. For the joint anchor+H II galaxies sample, under a flat $\Lambda$CDM model, we obtain $h=0.725\pm0.040$ and $\Omega_m=0.308^{+0.043}_{-0.053}$. Allowing a constant dark-energy equation of state yields $w_0=-0.96^{+0.53}_{-0.21}$, while a CPL parametrisation gives $w_0=-0.92^{+0.57}_{-0.34}$ and $w_a=-0.48^{+0.60}_{-1.50}$, all consistent, within the uncertainties, with concordance cosmology. These results demonstrate that H II galaxies offer a viable, fully independent tracer of the expansion history across almost the entire age of the Universe, opening a new avenue for testing $\Lambda$CDM and dark-energy evolution well beyond reionisation.

arXiv:2607.14251v1 Announce Type: new Abstract: Metal-enriched white dwarfs provide a unique insight into the composition of exoplanet interiors. These stars accrete the debris of disrupted planetary bodies, and hence, measuring the stellar parameters and photospheric abundances yields the bulk compositions of the parent bodies. At present, over 1750 debris-accreting white dwarfs are known, but just a few dozen are sufficiently enriched to allow a detailed abundance study. Here we report the analysis of 12 highly metal-enriched white dwarfs observed within the Data Release~1 of the Dark Energy Spectroscopic Instrument (DESI). We characterised their stellar parameters and photospheric metal abundances and we identified between three and ten different elements in their optical spectra, including most of the rock-forming species: O, Mg, Si, Ca and Fe. We conclude that the accreted bodies broadly resemble compositions found within the inner Solar System such as primitive meteorites, processed material or planetary cores. Six of the systems allowed a more thorough analysis: four of the parent bodies are composed of dry rock-forming elements; and two of them of something akin to a water-rich planetesimal. Thus, this study establishes DESI as a potent survey for identifying metal-rich targets, yielding reliable compositions of accreted exoplanetary material.

arXiv:2601.06255v2 Announce Type: replace Abstract: We present a search for and characterization of ionized outflows in 15 star-forming systems at $z\sim3-9$ with no evidence of Active Galactic Nuclei (AGN), observed with JWST/NIRSpec IFU as part of the GA-NIFS program. The targets often show satellites and complex substructure, from which we isolate 40 galaxies/regions. The sample probes the high-mass end of the galaxy population, with most sources having log$_{10}$~(M$_\star$/M$_\odot$)=$9.5-11$, extending previous studies on high-z star formation driven outflows that mainly focused on lower-mass galaxies. Using the [OIII]5007 and H$\alpha$ emission lines, we identify broad kinematic components consistent with galactic outflows in 14 galaxies/regions. We find that the outflowing gas is more dust attenuated (by $A_{\rm V}$$\sim0.59$ mag on average) and metal-enriched (0.13 dex) than the interstellar medium (ISM) of the host galaxies, but its velocities are insufficient to escape the galaxies and reach the circumgalactic medium, suggesting that outflows mainly redistribute dust and metals around their hosts. The outflows identified in this study display velocity dispersions within $\sigma_{\rm out}=130-340$~km~s$^{-1}$ and outflow velocities $v_{\rm out}=170-600$~km~s$^{-1}$, and, when combined with less luminous and less massive star-forming galaxies from previous works, reveal a statistically significant correlation between $v_{\rm out}$ and star formation rate (SFR). The typically low mass-loading factors ($\eta=\dot{M}_{\rm out}/SFR$$\leq1$, in 9 out of 14 the outflows) indicate that these outflows do not strongly suppress star formation. Overall, our results suggest that ejective feedback via ionized outflows is inefficient in massive, luminous star-forming galaxies within the first 2 Gyr of the Universe.

arXiv:2607.11438v2 Announce Type: replace Abstract: Mutual coupling between antennas has emerged as the dominant direction-dependent corruption in dense aperture arrays, imprinting pronounced sub-MHz spatial and spectral structure that compromises the time-gating and foreground-separation strategies used to isolate the faint 21-cm signal. In this work, we introduce \textit{Direct Primary Beam Correction}, a domain-agnostic framework for reconstructing the far-field radiation pattern relative to an arbitrary reference via a regularised, direction-weighted linear inversion of stacked Jones matrices, thereby enabling the removal of direction-dependent distortions such as mutual coupling. Using full-wave electromagnetic simulations of SKA-Low, we demonstrate that this framework reconstructs the radiation pattern down to the numerical noise floor within a suitably conditioned field of view, with the reconstruction accuracy governed by the regularised inversion and the fidelity of the underlying beam model. Applying the framework to a simulated 4-hour observation of the EoR0 field in the $122$--$134$~MHz band, we identify two principal implications for 21-cm power-spectrum analysis. First, restricting the correction to the main lobe and near sidelobes is inadequate: chromatic grating lobe contributions, whether left insufficiently or entirely uncorrected, continue to contaminate the EoR window. Second, mutual-coupling-induced contamination is temporally coherent and, being anchored to the fixed array geometry, does not average down across snapshots as the EoR field is tracked. Direct primary beam correction, therefore, provides a computationally efficient means of mitigating mutual coupling; however, robust recovery of the EoR window necessitates either full-sky correction or explicit separation of main-beam and sidelobe contributions prior to power-spectrum estimation.

arXiv:2606.10011v2 Announce Type: replace Abstract: In March 2026, the Einstein Probe (EP) discovered its most nearby (z = 0.0343) Fast X-ray Transient (FXT), EP260321a, the first EP FXT to provide a strong match to expectations for X-ray "shock breakout'" (SBO) emission. Here, we present our multi-wavelength follow-up campaign of EP260321a and its broad-line Type Ic (Ic-BL) supernova (SN) counterpart, SN2026gzf. We show that our radio follow-up extending over 5.8 - 54.5 days post-FXT rules out an on-axis jet counterpart of isotropic-equivalent kinetic energy $E_{K} \gtrsim 10^{49}$ erg for circumburst densities $n > 10^{-2}~{\rm cm}^{-3}$ and assuming microphysical parameters $\epsilon_e = \epsilon_B = 0.1$. Our radio data also constrains a median mass-loss rate of $\dot{M} \lesssim 1.2 \times 10^{-5} M_{\odot}~{\rm yr}^{-1}$ for a Wolf-Rayet progenitor. In addition, we derive SN2026gzf's properties, including $^{56}$Ni mass, diffusion timescale, and expansion velocities, from our $\sim$nightly-cadence optical data and compare them with those of optically discovered Type Ic-BL SNe, finding that SN 2026gzf is well within the 90\% confidence interval across all properties. We further fit SN2026gzf's light curve and determine that combined emission from both interaction with CSM and $^{56}$Ni radioactive decay provides the best fit with plausible model parameters. Finally, using the rate of Ic-BL SNe from the ZTF Bright Transient Survey and assuming all Type Ic-BL SNe produce EP260321a-like FXTs, we infer an expected rate of EP-detected SBOs of 4.4 - 16 year$^{-1}$. This is inconsistent at the 90% confidence level with current EP detection rates, potentially indicating that most Type Ic-BL SNe produce less luminous X-ray SBO signals compared to EP260321a.

arXiv:2607.12018v1 Announce Type: new Abstract: The JADES Transient Survey (JTS) identified 83 supernova (SN) candidates in the JADES Deep Field, a $\sim$25 arcmin$^2$ region with deep ($\sim$30 mag) multi-band, multi-epoch JWST/NIRCam coverage. We use this sample to derive the first volumetric core-collapse (CC) SN and Type Ia (SN Ia) rates in the $z$$\sim$2-5 range. Many of these SNe are photometrically classified from single-epoch photometry (i.e., single spectral energy distributions (SEDs)), so we simulate and classify $\sim$23,000 CC SN and SN Ia mock SEDs over 0.7$\leq$$z$$\leq$5 to quantify single-SED classification accuracy as a function of redshift. We report consistent rates for two samples: (1) the full JTS sample, including single-SED classifications, and (2) the "gold" sample, restricted to sources classified spectroscopically or with multi-epoch light curves. In units of 10$^{-4}$ CC SNe yr$^{-1}$ Mpc$^{-3}$, the full sample CC SN rates are 6.2$^{+2.2}_{-1.7}$ at 2.06$\leq$$z$$<$2.78 and 4.1$^{+1.5}_{-1.1}$ at 2.78$\leq$$z$$\leq$5.06, broadly consistent with the expectations from the galaxy luminosity-based measurements of the cosmic star formation rate density. Our full sample rates tentatively exhibit the predicted decline beyond cosmic noon, providing the first direct observational indication of this behavior. A companion paper, C. Vassallo et al., presents a more detailed comparison. We measure a full sample SN Ia rate of 0.3$^{+0.3}_{-0.2}$$\times$10$^{-4}$ SNe Ia yr$^{-1}$ Mpc$^{-3}$ at 1.92$\leq$$z$$<$3.60. Future high-$z$ SN surveys with JWST and the Roman Space Telescope will expand these samples and provide more robust constraints on SN rates in the high-$z$ Universe.

arXiv:2607.08824v1 Announce Type: new Abstract: Cosmic dust is a key regulator of galaxy evolution, but its build-up and survival in the first billion years remain poorly constrained. We present a systematic analysis of dust in the thesan-zoom suite of radiation-hydrodynamical zoom-in simulations, which self-consistently model dust formation, growth, destruction, and its coupling to radiative transfer in galaxies at $z \geq 3$, a multi-phase ISM and bursty star formation histories. The simulated galaxies reproduce the observed trends of dust-to-gas and dust-to-metal ratios with gas metallicity, while showing a dust deficit at high specific star-formation rates. They also broadly match observed dust temperatures and UV-IR spatial offsets. We find that dust and its properties are strongly time-variable and tightly linked to bursty star formation, with short-lived IR-bright phases (median duration of $20.3^{+2.3}_{-2.4}$ Myr) and longer dust-poor phases, naturally producing a correlation between dust temperature and distance from the star-forming main sequence. The predicted attenuation at $1500$ \r{A} is low compared to observations, even when including unresolved dust through post processing, indicating that a mechanism able to shield dust from strong feedback events is necessary to reconcile our galaxy formation model with observations. In our model, bursty star formation prevents the survival of large dust reservoirs ($M_{dust} / M_{star} \geq 10^{-3}$) over a significant fraction of cosmic time. This implies that bursty star formation can produce the observed overabundance of UV-bright galaxies at $z \geq 10$ only if it rapidly settles down by $z \sim 8$ (where large dust reservoirs are detected). It is also possible that our models lack physical ingredients or emergent phenomena that aid the survival of dust. Future observations of high-redshift dust will be key to diagnose the physical mechanism at play in the first galaxies.

arXiv:2607.08090v1 Announce Type: new Abstract: Fast X-ray transients (FXTs) are a diverse class of high-energy suggested origins, ranging from stellar explosions to compact object mergers. The Einstein Probe (EP) satellite discovers approximately 100 FXTs per year. We aim to constrain the physical origin of EP250905a. We analyze X-ray, optical, near-infrared (NIR), and radio temporal and spectral properties of EP250905a. In addition, we assess the possible role of weak gravitational lensing in shaping its observed characteristics. EP250905a fades rapidly in X-rays, and we detect no NIR or radio emission, but we detect early optical emission that rapidly fades beyond the detection limits. Two nearby galaxies are identified for which we derive spectroscopic redshifts of $z=0.374$ (G1) and $z=2.714$ (G2). Our analysis favors G2 as the host of the FXT EP250905a. The angular separation of 2.56\arcsec\, between the FXT's optical counterpart and the center of the G1 galaxy suggests the emission of the FXT might be moderately magnified by lensing effects ($\mu\approx3.9$) given the inferred Einstein radius of G1 ($\theta_E\approx1.9$ arcsec). The data are best explained as an afterglow from a mildly off-axis structured jet at $z=2.714$, providing a consistent broadband interpretation that also allows for weakly lensed emission of EP250905a.

arXiv:2507.13905v2 Announce Type: replace Abstract: We systematically extend the framework of galaxy bias renormalization to two-loop order. For the minimal complete basis of 29 deterministic bias operators up to fifth order in the density field and at leading order in gradient expansion we explicitly work out one- and two-loop renormalization. The latter is provided in terms of double-hard limits of bias kernels, which we find to depend on only one function of the ratio of the loop momenta. After including stochasticity in terms of composite operator renormalization, we apply the framework to the two-loop power spectrum of biased tracers and provide a simple result suitable for numerical evaluation. In addition, we work out one- and two-loop renormalization group equations (RGE) for deterministic bias coefficients related to bias operators constructed from a smoothed density field, generalizing previous works. We identify a linear combination of bias operators with enhanced UV sensitivity, related to a positive eigenvalue of the RGE. Finally, we present an analogy with the RGE as used in quantum field theory, suggesting that a resummation of large logarithms as employed in the latter may also yield useful applications in the study of large-scale galaxy bias.

arXiv:2501.16292v3 Announce Type: replace Abstract: Inhomogeneities along the line of sight in strong gravitational lensing distort the images produced, in an effect called shear. If measurable, this shear may provide independent constraints on cosmological parameters, complementary to traditional cosmic shear. We model 23 strong gravitational lenses from the Sloan Lens ACS (SLACS) catalogue with the aim of measuring the line-of-sight (LOS) shear for the first time. We use the 'minimal model' for the LOS shear, which has been shown to be theoretically safe from degeneracies with lens model parameters, a finding which has been confirmed using mock data. We use the dolphin automated modelling pipeline, which uses the lenstronomy software as a modelling engine, to model our selected lenses. Across the 23 lenses, we measure the LOS shear with a mean magnitude of $0.056 \pm 0.013$. Neglecting the post-Born correction to the potential of the main deflector due to foreground shear leads to a propagation of degeneracies in the LOS shear measurement with other lens model parameters, and the inclusion of an octupole moment in the lens mass profile does not lead to shear measurements that are in better agreement with the expectations from weak lensing.

arXiv:2607.06680v1 Announce Type: cross Abstract: While vacuum decay is traditionally described by Euclidean instanton methods, lattice simulations enable real-time modeling of dynamical observables relevant to cosmology and upcoming cold-atom analog experiments. We investigate the relationship between these approaches by extracting ensemble-averaged bubble profiles from zero-temperature simulations of a relativistic scalar field. Our observed profiles differ markedly from the bare Coleman bounce and classical thermal predictions. However, we find that instanton solutions in an appropriately renormalized potential reproduce both the measured profiles and their dependence on the UV cutoff, and predict decay rates consistent with simulations across the parameter range considered. The fact that a single renormalized Euclidean object captures these independent observables provides strong evidence that renormalization accounts for the discrepancy between the two formalisms, and establishes a quantitative link between instanton predictions, lattice simulations, and forthcoming empirical tests of vacuum decay.

arXiv:2511.05484v2 Announce Type: replace Abstract: We revisit the stochastic, or noise, contributions to the galaxy density field within the effective field theory (EFT) of large-scale structure. Starting from the general, all-order expression of the EFT partition function, we elucidate how the stochastic contributions can be described by local nonlinear couplings of a single Gaussian noise field. We introduce an alternative formulation of the partition function in terms of such a noise field, and derive the corresponding field-level likelihood for biased tracers. This noise-field formulation can capture the complete set of stochastic contributions to the galaxy density at the field level in a normalized, positive-definite probability density which is suitable for numerical sampling. We illustrate this by presenting the first results of EFT-based field-level inference with non-Gaussian and density-dependent stochasticity on dark matter halos using LEFTfield.

arXiv:2606.30504v2 Announce Type: replace Abstract: The bright-siren measurement of the Hubble constant from GW170817 (Abbott et al. 2017) assumes that switching from a volumetric to a uniform-in-$d_L$ luminosity-distance prior can be implemented by post-hoc reweighting of the baseline samples, rather than by re-running the inference under the target prior. Using a GPU-native heterodyned nested sampling pipeline that completes the full $n_{\rm live}=5000$ analysis in about 13 min on a single A100, we recompute the GW170817 $H_0$ posterior under four prior variants for the modern aligned-spin tidal waveform IMRPhenomXAS_NRTidalv3. Switching from the volumetric to a uniform-in-$d_L$ distance prior raises the high-tail probability $P(H_0>120\,\mathrm{km/s/Mpc})$ from 0.017 to 0.159 when imposed during sampling and shifts the weighted-median $H_0$ from 77.6 to 87.6 km/s/Mpc, while the binned MAP stays at 70.5 km/s/Mpc: both the tail and the bulk move under a change of prior that leaves the mode in place. Post-hoc reweighting of the baseline samples to the same target prior recovers only $P=0.041$ in the tail, approximately 17% of the directly sampled shift. The three prior variants that carry an independent nested sampling evidence agree to $\Delta\ln Z\lesssim 1.8$, so the data show at most a weak preference among the distance priors; the tail and bulk shifts are therefore properties of the prior, not a data update. Targeted mode-isolated runs reveal a $(d_L,\iota)$ bimodality whose high-$H_0$, low-$d_L$ branch (Mode B; $|\ln\mathcal{B}_{\rm B/A}|<1$) the volumetric prior assigns negligible mass: this is the mechanism behind the reweighting deficit. The reweighted posterior has a lower effective sample size than the baseline, independently flagging the coverage failure. The runtime budget makes full-sample prior-sensitivity reruns the default robustness tool for bright-siren cosmology, replacing post-hoc reweighting.

arXiv:2607.05523v1 Announce Type: new Abstract: Despite the growing number of compact active galactic nuclei (AGN) at $z>4$ discovered by JWST, their formation and evolution remain poorly understood. This paper investigates the large-scale environment of GN-77652, a compact AGN at $z=5.229$ observed as part of the JWST NIRSpec IFU Large Program BlackTHUNDER and complemented by deep multi-band NIRCam imaging. GN-77652 lies in close proximity to a 12 kpc-long filament composed of multiple sources at $z\simeq5.23$, spanning a remarkable range in stellar masses ($M_{\star}=0.7-13 \times 10^8$ ${M_\odot}$), gas phase metallicities (12$+$log(O/H) $=$ 7.6-8.5) and star formation rates (SFR $=0.4-6$ ${M_\odot}$ yr$^{-1}$). The [OIII]$\lambda$5007 kinematics reveals a smooth large-scale velocity gradient centred on the central, massive ($M_{\star}\simeq1.1\times10^9$ ${M_\odot}$) and metal rich ($Z\sim0.6$ $Z_{\odot}$) system of the group. In this source, only 2.4 kpc (projected) from GN-77652, [OIII]$\lambda$4363 line diagnostics provide possible evidence for a second AGN. GN-77652 exhibits a shallow ($-30$ to $+20$ km s$^{-1}$) velocity gradient that is consistent with disk rotation according to dynamical modelling. The Lyman-Werner radiation field produced by the filament is too weak for the black hole (BH) in GN-77652 to have formed recently via direct collapse. However, the required conditions may have existed at earlier epochs, or alternative scenarios (e.g. a recoiling BH ejected from the filament) could also be plausible. The whole system is expected to coalesce in $150-440$ Myr, also motivating an exploration of its future evolution through toy-model extrapolations and numerical simulations. Our analysis suggests that the compact AGN appearance of GN-77652 represents a transient evolutionary phase, consistent with the apparent decline with redshift in number density of compact AGN identified with JWST.

arXiv:2607.05509v1 Announce Type: new Abstract: Globular clusters exhibit large star-to-star variations and anticorrelations in their light element abundances that are commonly interpreted in terms of in-cluster self-enrichment, in which ejecta from early-forming cluster stars pollute the gas from which later stars form over millions of years. Yet proposed self-enrichment scenarios suffer from a severe mass-budget problem or invoke exotic stellar populations. Using cosmological radiation-hydrodynamic simulations with a standard chemical enrichment model, we identify a population of giant molecular clouds whose internal abundance patterns reproduce several key globular cluster signatures: large light-element abundance spreads and nitrogen-oxygen anticorrelations at nearly constant iron abundance. These clouds form at the restart of star-formation activity after an earlier starburst, where previously ejected oxygen-rich gas collides with nitrogen-rich galactic gas, and are sites of dense star-cluster formation. In this picture, the chemical abundance patterns of globular clusters need not require extended in-cluster star formation, but can be inherited at birth from chemically structured interstellar gas shaped by the baryon cycle. Globular clusters therefore provide a fossil record of chemical enrichment and gas flows in high-redshift galaxies.

Explaining the passage of time has been a gnarly problem in physics basically forever, but physicist and computer scientist Stephen Wolfram has a radical proposal for where it comes from. He discussed his ideas on time – and what they mean for free will – with reporter Leah Crane

arXiv:2607.01048v2 Announce Type: replace Abstract: The Square Kilometre Array (SKA), with its unprecedented sensitivity, frequency coverage, and large collecting area, is poised to revolutionize our understanding of the Cosmic Dawn (CD) and Epoch of Reionization (EoR) epochs marking the formation of the first luminous sources and the subsequent reionization of the intergalactic medium (IGM). However, detecting the faint redshifted 21-cm signal from neutral hydrogen remains one of the foremost challenges in observational cosmology, as it is buried beneath bright foregrounds from Galactic synchrotron radiation, free-free emission, and extragalactic point sources that are 4-5 orders of magnitude stronger than the cosmological signal. In this chapter, we highlight the key components and characteristics of these foregrounds and review ongoing efforts to model, characterize, and mitigate them. We emphasize how the SKA-Low AA* configuration, through its optimized array design, wide field of view, and improved calibration accuracy, enhances our capacity to suppress foreground contamination and recover the cosmological signal. The SKA Observatory Foreground Challenge plays a pivotal role in this effort by bringing together the global EoR/CD community to develop, compare, and validate foreground removal pipelines using realistic simulated datasets. Building on the experience of existing pathfinders such as LOFAR, MWA, and HERA, these collaborative initiatives are helping refine statistical and machine learning-based approaches for signal recovery. Together, these advancements are laying the groundwork for the SKA to probe the thermal and ionization history of the early Universe with unprecedented precision.

arXiv:2511.10499v2 Announce Type: replace Abstract: The 21-cm signal from neutral hydrogen traces the formation and evolution of early cosmic structures during the Cosmic Dawn and the subsequent Epoch of Reionization. However, the intrinsic faintness of the signal, as opposed to astrophysical foregrounds, poses a formidable challenge for its detection. Motivated by the recent success of machine learning based Gaussian Process Regression (GPR) methods in LOFAR and NenuFAR observations, we perform a Bayesian comparison among five GPR models to account for simulated 4-hour tracking observations with the SKA-Low telescope. The simulations incorporate the beam response of the telescope and include realistic radio sources and thermal noise from 122 to 134 MHz. A Bayesian model evaluation framework is applied to five GPR models to discern the most effective modelling strategy and determine the optimal model parameters. The GPR model with wedge parametrization ($\textit{Wedge}$) and its extension ($\alpha\textit{Noise}$) with noise scaling achieve the highest Bayesian evidence of the observed data and the least biased 21-cm power spectrum recovery. The $\textit{Wedge}$ and $\alpha\textit{Noise}$ models also forecast the best local power-spectrum recovery, demonstrating fractional differences of $0.10\%$ and $-0.24\%$ respectively, compared to the injected 21-cm power at $k = 0.32\ \mathrm{h\ cMpc}^{-1}$. We additionally perform Bayesian null tests to validate the five models, finding that the two optimal models also pass with the remaining three models yielding spurious detections in data containing no 21-cm signal.