Peer Revenue Past citations are no guarantee of future returns

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

Physics Astrophysics

astro-ph Watch Astrophysics

Thursday 9 July

High Energy Astrophysical Phenomena

A cooler look at the environment of Cygnus X-1: Searching for dynamical interactions within cold molecular gas

Pau Bosch-Cabot, Alexandra J. Tetarenko, Valentí Bosch-Ramon et al. (+7 more)

We present IRAM–30m observations aimed at identifying potential outflow-interstellar medium interaction sites in the vicinity of the black hole X-ray binary Cygnus X–1, which displays persistent relativistic jets and a prominent stellar wind. Using this dataset, we construct molecular line emission maps, identifying a never before seen molecular structure potentially linked to X-ray binary-driven feedback. This structure, surrounding the source, exhibits properties consistent with an interaction powered primarily by the stellar wind of the massive stellar companion and further sculpted by the relativistic jets. Our results indicate that multiple outflow mechanisms (stellar winds and relativistic jets) may simultaneously be shaping the interstellar medium around Cygnus X–1, and that molecular line imaging can help to disentangle complex feedback processes in environments where multiple outflows take place.
Read the full abstract

All authors Pau Bosch-Cabot, Alexandra J. Tetarenko, Valentí Bosch-Ramon, James C.A. Miller-Jones, David M. Russell, Sara E. Motta, Pikky Atri, María Díaz-Trigo, Isabella Mariani, Steve Prabu

Solar and Stellar Astrophysics

Developing Machine Learning Models of Subgrid Turbulent Transport for Quiet Sun 3D Radiative Hydrodynamic Simulations

Rimsha Hameed Syeda, Dustin Kempton, Viacheslav Sadykov et al. (+2 more)

Numerical modeling of solar plasma dynamics is affected by the resolution of the computational grid. This often requires the estimation of subgrid processes related to the small-scale flow turbulence, as these processes play a critical role in momentum transport and energy dissipation. In this work, we investigate the use of deep learning techniques as surrogate models for subgrid turbulent transport in realistic hydrodynamic simulations of the quiet Sun. We describe the development of a 3D Convolutional Neural Network (CNN) to capture spatial dependencies in 3D velocity fields, leveraging different activation functions, as well as different architectural designs. We specifically focus on the prediction of Reynolds stress tensor components. The resultant model integrates velocity vector components and scalar features, such as plasma density, to enhance prediction accuracy. We compare the 3DCNN model to other types of models, such as a Multilayer Perceptron (MLP) and physics-based Gradient and Smagorinsky models, and show that the final model design reconstructs the Reynolds stress tensor components more accurately. Specifically, a 3DCNN model achieves an average improvement of ~31% on diagonal components and ~8% on the off-diagonal components of the stress tensor. Additionally, we show that applying a logarithmic data transformation of the target stress tensor components, to handle heavily skewed data, improves model performance. Results demonstrate the potential of deep learning, particularly CNNs, to approximate Reynolds stress tensor components for the upper solar convection zone and lower atmosphere, making them a viable candidate for modeling subgrid processes and a promising alternative to traditional turbulence models.
Read the full abstract

All authors Rimsha Hameed Syeda, Dustin Kempton, Viacheslav Sadykov, Irina Kitiashvili, Rafal Angryk

Astrophysics of Galaxies

Star Clusters in the Ultraviolet

Annapurni Subramaniam, Samyaday Choudhury, Richard de Grijs et al. (+5 more)

Ultraviolet (UV) observations provide a powerful window into the hot and evolved stellar populations that shape the structure, evolution and integrated light of star clusters. Because UV wavelengths are highly sensitive to massive main-sequence stars, blue straggler stars (BSS), extreme-horizontal branch (HB) stars, post-HB stars, interacting binaries and compact remnants, they probe key evolutionary processes that are inaccessible at optical and infrared wavelengths. This review synthesises five decades of UV studies of star clusters across the Milky Way, the Magellanic Clouds (MCs) and nearby galaxies, drawing on results from early space missions, wide-field surveys, and high-resolution imaging. In Galactic open clusters, UV studies have revealed compact companions$-$including white dwarfs, hot subdwarfs and stripped stars$-$and have established the mass-transfer origins of BSS, Blue Lurkers and yellow stragglers. In globular clusters, UV imaging has identified multiple stellar populations through UV-sensitive molecular bands, probed helium enrichment and mapped HB morphologies. Wide-field UVIT surveys have extended HST studies with homogeneous catalogues of HB and post-HB stars across entire clusters. In the MCs, UV observations have transformed our understanding of multiple populations, rotation-driven extended main-sequence turn-offs and the UV-dim phenomenon, while spectroscopic surveys have constrained massive-star evolution, stellar winds and binarity at low metallicity. UV mapping of the Magellanic Bridge has revealed ongoing massive-star formation in low-density tidal environments. Beyond the Local Group, UV studies of extragalactic clusters constrain star-formation histories, stellar feedback and population synthesis across galactic environments. Collectively, UV observations now form a cornerstone of star cluster astrophysics and will continue to do so with upcoming missions.
Read the full abstract

All authors Annapurni Subramaniam, Samyaday Choudhury, Richard de Grijs, Vikrant V. Jadhav, Chengyuan Li, Snehalata Sahu, Kaushar Vaidya, Li Wang

Earth and Planetary Astrophysics

Tentative detection of circularly polarized bursty radio emissions from the HD 189733 exoplanetary system using NenuFAR beamformed observations

Jake D. Turner, Philippe Zarka, Jean-Mathias Grießmeier et al. (+4 more)

Observing auroral radio emission is one of the most promising methods for detecting exoplanetary magnetic fields, which provide valuable insights into planetary interiors, atmospheric properties, and potential habitability. The first hints of exoplanet auroral emission are starting to emerge. Recently, Zhang et al. (2025) reported a detection at 50 MHz of a circularly polarized bursty emission from the HD 189733 exoplanetary system using NenuFAR low-frequency imaging observations. The source of the emission is still unknown and may be caused by planetary auroral emissions, star-planet interactions, stellar activity, or the M-dwarf stellar companion. In this study, we analyze beamformed observations from NenuFAR of HD 189733 taken simultaneously during the previously detected burst. This dataset allows for an independent verification of the detected burst with a different backend and processing steps. Using the BOREALIS data reduction pipeline, we tentatively detect circularly polarized bursty emission ($\sim$10$\sigma$) from HD 189733 $\sim$1 hour before the burst found from the imaging observations. However, some uncertainty remains on whether our detected signal is astrophysical in nature due to excess correlated noise. Assuming an astrophysical origin, our observed characteristics are most consistent with a planetary origin, but stellar emission cannot be completely ruled. Therefore, more low-frequency radio observations are needed to confirm the astrophysical nature of our signal and to search for periodicity in the radio signal from HD 189733 to determine the true cause of the emission. These observations are ongoing. Our study highlights the power of simultaneous beamformed and imaging observations in the search for radio emission from exoplanets.
Read the full abstract

All authors Jake D. Turner, Philippe Zarka, Jean-Mathias Grießmeier, Corentin K. Louis, Xiang Zhang, Emilie Mauduit, Tomoki Kimura

Cosmology and Nongalactic Astrophysics

Chromatic Effects Across the Roman Focal Plane: Implications for Supernova Photometry and Measurements of Cosmological Parameters

Rujuta Purohit, Dillon Brout, Daniel Scolnic et al. (+14 more)

Calibration uncertainties are the leading systematics in cosmological analyses using Type Ia supernovae (SNe Ia). For the Nancy Grace Roman Space Telescope (Roman), we quantify the impact of chromatic effects on SNe Ia photometry and derived cosmological parameters, using simulated light curves from the High-Latitude Time Domain Survey. We investigate two sources of wavelength-dependent bias: focal plane array (FPA)-dependent wavelength shifts arising from spatial variations across Roman's 18 detectors, and coherent wavelength shifts corresponding to the measured $0.06\%$ uncertainty in absolute filter wavelength calibration. Using simulated SNe Ia light curves, we find that the FPA-dependent shifts – which range from +6 to -80 $\rm \AA$ introduce a redshift-dependent distance modulus bias that, if left uncorrected, propagates to $\Delta w_0 \sim -0.06$ and $\Delta w_a = -0.236$, which are larger than the forecast statistical uncertainties of $\sigma_{\rm stat, w_0} = 0.025$ and $\sigma_{\rm stat, w_a} = 0.114$, rendering the survey systematics-limited. We probe the impact of chromatic effects by employing detector-specific filter curves that recover unbiased cosmological constraints; to remain below the statistical noise floor, FPA wavelength shifts must be characterized to within 20%. In contrast, a coherent 0.06% offset in filter wavelength calibration – ranging from -3 to -11 $\rm \AA$ – produces negligible redshift-dependent bias, with a minimal spread in $w_a$ ($\Delta w_a = -0.0004, \sigma_{w_{a,\rm sys}} = 0.114)$, demonstrating that the achieved pre-launch calibration precision is sufficient for this systematic to remain subdominant. Our results establish that chromatic effects are a required component of SN Ia cosmology with Roman.
Read the full abstract

All authors Rujuta Purohit, Dillon Brout, Daniel Scolnic, Richard Kessler, Jillian Paulin, Stefano Casertano, Nora F. Sherman, Rebekah Hounsell, Eric R. Switzer, Benjamin M. Rose, Masao Sako, Lauren Aldoroty, Kene Anumba, Maria Acevedo, David Rubin, Russell E. Ryan Jr., The Roman Supernova Cosmology Project Infrastructure Team

High Energy Astrophysical Phenomena

Studying the Variable Continuum and Emission Line Spectrum of 4U 1700-377 using NuSTAR's Stray Light

Colin J Bourque, McKinley C Brumback, Brian W Grefenstette et al. (+4 more)

The high-mass X-ray binary 4U 1700-377 shows strong variation in brightness on timescales of hundreds of seconds due to accretion from a highly clumped stellar wind from the companion. Using two focused observations and five stray light observations from NuSTAR, we are able to expand the baseline of observations of the source and track spectral parameters throughout the source's variability. The focused NuSTAR observations confidently detect excess emission lines at energies above Fe K$\alpha$, but are unable to universally constrain the nature of these lines independently of model choice. Strong Fe K$\alpha$ fluorescence is ubiquitous in the stray light data, but these observations lack sufficient signal-to-noise to further measure additional emission lines. We discuss the statistics of a claimed cyclotron resonant scattering feature as seen in the stray light spectra, but find no conclusive evidence for its existence. Finally, we investigate two instances of sustained low-flux states of up to 60 ks. When this occurs, luminosity can fall by almost an order of magnitude and the high-energy continuum softens significantly without corresponding to increased absorption. We discuss the manner in which the changing shape of the continuum during these intervals may show accretion during an extended rarefied interval of stellar wind, or a possible change of the dominant accretion regime.
Read the full abstract

All authors Colin J Bourque, McKinley C Brumback, Brian W Grefenstette, Douglas J K Buisson, Renee M Ludlam, Guglielmo Mastroserio, Steven Rossland

Instrumentation and Methods for Astrophysics

Lessons learned from ongoing coordination between NRAO/GBO and LEO NGSO satellite constellations

Bang D. Nhan, Christopher G. De Pree, Sheldon Wasik et al. (+3 more)

With the increased demand for spectrum usage in recent years, particularly with the expansion of NGSO satellite systems for broadband internet and cellular services, radio astronomy observatories located at remote sites have been faced with the challenge of increasingly detrimental RFI exposure from ubiquitous satellite downlinks. With the help of the U.S. NSF, NRAO and the radio astronomy community have spearheaded collaboration efforts with the commercial NGSO operators to develop different coordination techniques to alleviate the impacts of these downlink signals in radio astronomy scientific observations. This paper summarizes the ongoing coordination efforts, spectrum interference avoidance schemes, and a case study on community participation of the Operational Data Sharing (ODS) system between RAS and NGSO operators. We highlight some of the essential lessons learned from these collaborations and potential application of the ODS system to other potential spectrum stakeholders.
Read the full abstract

All authors Bang D. Nhan, Christopher G. De Pree, Sheldon Wasik, Aaron Lawson, Daniel Bautista, Frank Schinzel

Solar and Stellar Astrophysics

Observations of the Halo Star HD 177566

William V. Dixon and Pierre Chayer

We have analyzed archival FUV and optical spectra of the hot halo star HD 177566. The star has an effective temperature $T_{\rm eff} = 33{,}000 \pm 1000$ K, surface gravity $\log g = 3.79 \pm 0.11$, and helium abundance $\log N({\rm He})/N({\rm H}) = -0.86 \pm 0.05$. Abundances of 13 additional elements are consistent with those of other halo stars, save for carbon, which is underabundant by about 1 dex. The low-order hydrogen Balmer lines are not well reproduced by our models. The diffuse lines of He I are often broader than predicted, but the use of more recent line-broadening parameters significantly improves the fit. Scaling our best-fit model to the star's optical and near-IR magnitudes yields an extinction $E(B-V) = 0.095 \pm 0.005$, consistent with literature values, but the resulting model underpredicts the star's FUV flux by a factor of two. The star's effective temperature and luminosity ($\log L / L_{\odot} = 3.50 \pm 0.08$) place it on the post-AGB evolutionary tracks of a star that evolved from the red horizontal branch. Its low carbon abundance, $\log N({\rm C})/N({\rm O}) = -2.18 \pm 0.21$, indicates that it did not experience significant third dredge-up while ascending the AGB.
Read the full abstract

Instrumentation and Methods for Astrophysics

A Calibration Audit of a Gaia XP White-Dwarf Main-Sequence Binary Catalog: How Much the BP-Band Residual Contaminates at Its Bound and Above

Karan Akbari

A Gaussian-process classifier on Gaia DR3 XP spectra produced $\sim$30,000 white-dwarf main-sequence binary candidates (Li et al. 2025), with probabilities but no likelihood. The corrected spectra agree with external libraries to better than 2% below 400 nm and 1% redward (Huang et al. 2024), and that residual sits where a hot white dwarf also adds flux. We asked whether it contaminates the selection at its bound. It does, weakly. The spurious rate runs about 1.7 times the baseline there. Put into the blue band, 2% of total flux is a median 55% local excess across our MS templates, 27 times that number read locally. We inject a blue taper clipped to that bound, under the templates' published per-pixel errors, with a $\Delta\chi^2$ threshold standing in for the classifier. The spurious rate is $0.09\pm0.02$ across 20 seeds on a 0.053 baseline, 19 of 20 above it. The unclipped 2% taper, above the redward bound, gives $0.11\pm0.03$. At that taper the amortized posterior's 90% coverage falls to 0.85 from a clean 0.88 on the flat-model leg, with PIT uniformity rejected at $p<10^{-5}$ in all four seeds. The selection fails in bulk only above a 10-15% local excess, and the rate reaches 0.96 near 50%. A model-comparison gate certifies binarity only above a white-dwarf flux share near 0.05. At share 0.01, near the catalog's median of 0.0096, the $\Delta\chi^2$ preference inverts at a 20% local excess. Outside Huang's color-magnitude box, 40% of the catalog, the post-correction residual is not measured.
Read the full abstract

High Energy Astrophysical Phenomena

The SPOTLIGHT Pulsar Search Pipeline: A GPU-Accelerated FFT Approach

Jyotirmoy Das, Jayanta Roy, Nishant Pradeep Deo et al. (+20 more)

We present the pulsar search component of SPOTLIGHT (Survey for sPoradic radiO bursTs via a commensaL multI-beam Gpu-powered Hpc at the gmrT), a GPU-accelerated commensal backend operating at the upgraded Giant Metrewave Radio Telescope (uGMRT). While SPOTLIGHT is primarily designed for real-time detection and localisation of fast radio bursts (FRBs), it simultaneously records a subset of beamformed data products for periodicity searches without requiring dedicated telescope time. To process the large data volumes generated by the survey, we have developed a scalable FFT-based pulsar search pipeline that combines radio-frequency interference mitigation, GPU-accelerated dedispersion and periodicity searches, multi-beam candidate sifting, efficient folding and machine-learning classification. Using population synthesis and archival uGMRT observations, we estimate that a fully operational SPOTLIGHT survey with 160 PC and one IA beam could discover $\sim$ 450 new pulsars, probing both high-sky coverage and faint pulsars over three and a half years of commensal observations. The pipeline has been validated on GMRT Cycle 48 and 49 observations (i.e. April 2025 to Mar 2026), successfully re-detecting numerous known pulsars with a wide range of Period, DM and flux densities, and is currently operational for SPOTLIGHT commensal data processing. We describe the SPOTLIGHT observing system, pulsar survey design, search parameter space, candidate-selection strategy, current status, and future developments. SPOTLIGHT demonstrates the scientific potential of commensal pulsar surveys and serves as a pathfinder for real-time, large-scale pulsar and transient searches in the SKA era.
Read the full abstract

All authors Jyotirmoy Das, Jayanta Roy, Nishant Pradeep Deo, Karel Adámek, Wes Armour, Kshitij Bane, Jayaram Chengalur, Kenil Ajudiya, Chahat Dudeja, Sridhar Gajendran, Zhaocheng Gong, Param Joshi, Sanjay Kudale, Arpan Pal, Ujjwal Panda, Adityan S, Raghav Wani, Junhao Zhang, Deepak Bhong, Shelton Gnanaraj, Santaji N. Katore, Mekhala Muley, Harshavardhan Reddy

High Energy Astrophysical Phenomena · Cross-listed from nucl-th

Magnetised Dense Nuclear Matter in Neutron Stars: A Relativistic Mean-Field Study of the Equation of State

Mojdeh Banafsheh

We study the static response of dense neutron-star matter to a prescribed magnetic-field strength within a deliberately minimal and internally consistent relativistic mean-field (RMF) baseline. The model includes neutrons, protons, electrons, and muons in beta equilibrium and charge neutrality, with a uniform external magnetic field incorporated through Landau quantisation of the charged species. The equation of state is evaluated self-consistently at zero temperature, while moderate finite-temperature effects are estimated through leading-order degenerate Sommerfeld corrections. Magnetic pressure anisotropy is included through the Maxwell contribution in the no-magnetisation approximation. The purpose of this baseline is to isolate the hierarchy of magnetic-field effects before introducing additional microphysics or dynamical magnetic-field evolution. We compare the linear QHD-I parameterisation, used as a stiff benchmark, with the nonlinear GM1 model as a more realistic reference. The results show that canonical magnetar-scale fields produce negligible changes in the bulk core equation of state, while visible Landau-quantisation and pressure-anisotropy effects emerge only as the field approaches the strongly quantising regime. The comparison between QHD-I and GM1 further shows that nuclear-model dependence dominates over static magnetic-field corrections up to the field strengths explored here. As an exploratory diagnostic, we use the isotropised equation of state to estimate the sensitivity of ordinary TOV mass–radius sequences to the prescribed magnetic field. This should not be interpreted as a fully anisotropic magnetised-star calculation. Anomalous magnetic moments, hyperons, quark degrees of freedom, and dynamical magnetic-field evolution are left for future extensions.
Read the full abstract

Astrophysics of Galaxies

mitigating the overcooling problem with sink-based bursty star formation in a high-z dwarf galaxy

Cheonsu Kang, Taysun Kimm, Daniel Han et al. (+4 more)

Star formation is a fundamental driver of galaxy evolution, yet many galaxy formation models still fail to regulate it realistically, allowing gas to collapse too efficiently and overproduce stars. To investigate a possible solution to this overcooling problem, we perform cosmological zoom-in radiation-hydrodynamics simulations of a dark matter halo reaching $10^{10} M_\odot$ at $z=6$, adopting two distinct star formation models: a Schmidt-type model, in which star formation criteria and efficiency per free-fall time are tied to local gravo-thermo-turbulent conditions, and a sink-based model, in which star formation is governed by local gas inflows. The sink-based model naturally produces bursty star formation through rapid accretion onto young sink particles embedded in strongly convergent gas flows. The resulting intense radiation ionizes and disperses star-forming clumps through photoionization heating before the first supernova explodes. Consequently, supernovae occur in lower-density environments, imparting greater terminal momentum and driving stronger galactic outflows. In contrast, star formation within individual gas clumps is less efficient in the Schmidt-type model, because individual star formation events locally modify cell conditions, temporarily suppressing subsequent star formation and lowering the degree of burstiness. Relative to the Schmidt-type model, the sink-based model yields a total stellar mass lower by a factor of $\sim3$ and a Lyman continuum escape fraction higher by a factor of $\sim10$ by $z=6$. The bursty model drives stronger metal-enriched outflows and suppresses excess central star formation, exhibiting better agreement with JWST observations in gas-phase metallicity and galaxy size. Our results suggest that bursty star formation is a key mechanism for enhancing feedback and alleviating the overcooling problem in galaxy formation simulations.
Read the full abstract

All authors Cheonsu Kang, Taysun Kimm, Daniel Han, Maxime Rey, Fred Thompson, Martin P. Rey, Harley Katz

Cosmology and Nongalactic Astrophysics · Cross-listed from hep-ph

Dark Phoenix: dark matter relic from its own decay

Debasish Borah, Satyabrata Mahapatra, Indrajit Saha et al. (+2 more)

We propose a novel mechanism for generating correct relic of dark matter (DM) which otherwise gets thermally overproduced from the conventional freeze-out mechanism. The mechanism, dubbed as Dark Phoenix, relies on a transient decay window for DM after its freeze-out which brings its relic within observed limits. Due to finite-temperature effects on masses, DM $\psi$ becomes heavier than its dark sector partner $\phi$ in this window allowing it to decay. The decay of DM then stops after $\phi$ gets a sudden jump in its mass from a first-order phase transition (FOPT) driven by another scalar $\eta$. The dark sector partner $\phi$, assumed to be a charged scalar, undergoes sufficient pair annihilation during this epoch such that its late decay into DM does not overproduce the latter again. While direct-detection rates of DM remains suppressed due to small couplings, the charged scalar $\phi$ can have interesting signatures like long-lived charged track at colliders. The associated FOPT can also lead to observable gravitational waves at future experiments like LISA.
Read the full abstract

All authors Debasish Borah, Satyabrata Mahapatra, Indrajit Saha, Narendra Sahu, Vicky Singh Thounaojam

Astrophysics of Galaxies

Binary stars as dynamical tracers in globular clusters .II. Evolution of the radial distribution of the binary star fraction

J. Bruce, E. Vesperini, E. Dalessandro et al. (+5 more)

The distinct dynamical environments occupied by multiple stellar populations in globular clusters play a significant role in many aspects of cluster evolution, including the dynamics of binary stars. Recent observational analyses of Galactic globular clusters have revealed a bimodal radial trend in the binary fraction: the fraction of binaries is enhanced in both the central and outer regions of clusters, with a noticeable minimum in the cluster's intermediate regions. In this paper, we investigate the dynamical origin of this feature and the mechanisms responsible for preserving this bimodality for an extended portion of a cluster's lifetime. We utilize a suite of Monte Carlo simulations that follow the long-term dynamical evolution of both single-population and multiple-population globular clusters. We investigate how mass segregation and binary disruption function cohesively to produce a bimodal profile, and show that although single-population clusters can briefly generate a similar profile, the effect is weak and transient. Conversely, we show that the structural properties associated with the presence of multiple stellar populations significantly strengthen and preserve the bimodality. We also show that the effects of long-term dynamical evolution drive a broad relationship between a cluster's dynamical age and the radial location of the binary fraction minimum, which tends to migrate outward over time. Overall, our results strongly indicate that the typical multi-scale structure of multiple-population globular clusters (initially characterized by a dense and centrally concentrated subsystem of second-population stars embedded in a more extended first-population system) plays a key role in the origin of the observed bimodal profiles and they further demonstrate the analytical power of binary stars as dynamical tracers of globular cluster formation and dynamical evolution.
Read the full abstract

All authors J. Bruce, E. Vesperini, E. Dalessandro, M. Cadelano, F. R. Ferraro, B. Lanzoni, G. Beccari, C. Giusti

High Energy Astrophysical Phenomena

Hybrid Simulations of Proton Acceleration at Oblique High-$\beta$ Shocks

Yevhen Kylivnyk, Damiano Caprioli and Luca Orusa

Collisionless shocks in the intracluster and intergalactic medium (ICM/IGM) are expected to energize both electrons and ions. While electron acceleration is revealed by prominent radio emission, $\gamma$-ray emission from hadronic interactions remains undetected, suggesting that high-$\beta$ (ratio of thermal to magnetic pressure), low-Mach-number shocks cannot accelerate protons efficiently. We present three-dimensional hybrid simulations, in which ions are treated kinetically and electrons as a fluid, of quasi-perpendicular (magnetic obliquity $\vartheta = 80^\circ$) shocks with sonic Mach numbers $M_s \sim 3{-}15$ and plasma $\beta \gtrsim 15$, representative of cluster environments. We find that weak shocks ($M_s \lesssim 5$) fail to develop significant nonthermal populations, with cosmic ray (CR) acceleration efficiencies $\varepsilon_{\rm CR} \lesssim 0.1\%$. In contrast, stronger shocks ($M_s \gtrsim 10$) develop clear power-law tails with slopes $q \sim 4.0 $ and reach $\varepsilon_{\rm CR} \sim 3\%$. These results suggest that weak, oblique ICM shocks are generally unlikely to accelerate protons efficiently. However, reducing $\vartheta$ to $\sim 45^\circ$ leads to substantially higher acceleration efficiencies, indicating that magnetic obliquity plays a critical role in determining proton acceleration. Our findings provide a microphysical framework for interpreting radio relic observations, whose polarization suggests that electrons are accelerated at oblique shocks, and the absence of cluster $\gamma$-ray detections.
Read the full abstract

Astrophysics of Galaxies

Binary Stars as Dynamical Tracers in Globular Clusters .I. First Observations of Bimodal Spatial Distributions

M. Cadelano, E. Dalessandro, J. Bruce et al. (+7 more)

We present the first homogeneous study of the radial distribution of the binary fraction across the full extent of six Galactic globular clusters (GCs) spanning a wide range of dynamical ages, from dynamically young systems to core-collapsed clusters. We measured the radial variation of the binary fraction using a combination of deep optical HST observations and wide-field ground-based data. For the first time, we provide evidence that the binary fraction in GCs does not decrease monotonically with radius, as commonly assumed, but instead exhibits a bimodal distribution characterized by an excess in the outer regions. Specifically, the binary fraction displays a central peak, followed by a minimum at intermediate radii and a rising branch beyond approximately 1-2 half-light radii. The position of this minimum correlates with the cluster relaxation timescale, indicating that it is shaped by long-term dynamical effects of two-body relaxation driving the binary evolution, segregation, and disruption. The minimum radius also correlates with the A+ parameter derived from the radial distribution of blue straggler stars, an empirical indicator of the cluster dynamical age, further supporting the interpretation that this feature is due to internal dynamical processes. Numerical simulations presented in a companion paper show that such bimodal distributions naturally arise from the combined effects of binary disruption and mass segregation of the surviving binaries in clusters.
Read the full abstract

All authors M. Cadelano, E. Dalessandro, J. Bruce, E. Vesperini, F. R. Ferraro, B. Lanzoni, G. Beccari, C. Giusti, F. Cusano, D. Paris

High Energy Astrophysical Phenomena

Massive Perturbers and Transient Pickup Discs in Disc-Crossing Encounters in OJ 287-like Supermassive Black Hole Binaries

Debojyoti Garain, Jonathan Zrake and Zoltan Haiman

We study the hydrodynamical response of a massive black hole accretion disc punctured by a lower mass black hole on an inclined, eccentric orbit, as motivated by the quasi-periodic outbursts seen in the blazar OJ 287. Using three-dimensional smoothed particle hydrodynamics simulations, we explore how the secondary black hole's mass and orbital eccentricity, and the disc thickness and viscosity, affect the time variation of mass accretion onto both black holes. We find that disc-crossing events only lead to significant spikes in the primary black hole's accretion rate when the perturber is quite massive ($q \gtrsim 0.1$) and that such spikes are delayed by roughly a free-fall time ($\sim$ months for OJ 287) after the first disc-crossing. The orbital eccentricity, disc thickness, and viscosity can influence the amplitude and temporal structure of the response; in particular, higher eccentricity causes larger and less delayed delivery of gas to the primary, however the condition $q \gtrsim 0.1$ seems to be robust. We also show that when the secondary is this massive, it generally acquires a "pickup" disc which could produce its own luminous signature. For systems like OJ 287, the secondary pickup disc can contain $10^{2-3} M_\odot$, which if accreted over a timescale comparable to the orbital period can power near-Eddington secondary luminosities with thermal emission peaking in the UV/EUV.
Read the full abstract

Astrophysics of Galaxies

Testing the validity of backward orbit reconstruction for outer Local Group dwarf galaxies in IllustrisTNG-100

Priyanka Verma, Ekta Patel and Louis E. Strigari

Orbits of Milky Way (MW) and M31 satellites in cosmological simulations have been shown to match analytic orbits computed in fine-tuned models of the extended potential of the MW/M31 halos. Using a sample of Local Group (LG) analogs from the Illustris TNG100 simulations, we investigate whether the same is true for LG satellites near the periphery of the LG turnaround radius ($\sim 1\,{\rm Mpc}$), but outside the virial radii of the MW and M31. From both dark matter-only (TNG100-Dark) and full hydrodynamical (TNG100-1) simulations, we track the orbits of dwarf galaxy analogs and compare them with orbits resulting from an analytic model of the LG potential, which is dominated by the dark matter halos of the MW and M31, along with, where appropriate, their most massive satellites. We find that the reconstructed orbits of the selected outer LG satellites agree with the simulated orbits to within $\lesssim 10\%$ over the last 6 Gyr, with the scatter increasing with lookback time and reaching $\sim 40\%$ by 6 Gyr. The fact that orbits of outer LG dwarfs track the analytical models suggests that any excess mass in the outer regions of our LG analogs or the tidal field surrounding the LG does not affect the orbits of outer LG satellites. Residual differences between the analytic and simulated orbits are likely due to a combination of mass accretion for the satellites and MW/M31, the non-spherical nature of the MW/M31 potential, and the prescription for dynamical friction in the analytical models.
Read the full abstract

Cosmology and Nongalactic Astrophysics · Cross-listed from hep-ph

Seesaw reheating

Yann Mambrini

We introduce the Seesaw Reheating scenario, in which the inflaton transfers its energy to a long-lived intermediate scalar associated with the spontaneous breaking of lepton number before the Universe is reheated through its decay into right-handed neutrinos. As a result, the reheating temperature is no longer determined by the inflaton decay width but by the dynamics of the seesaw sector. We derive analytical solutions describing the complete reheating history, revealing two characteristic features of this scenario: the relativistic time dilation of the intermediate scalar, which suppresses its decay and delays the transfer of energy to the thermal bath, and its subsequent transition from a relativistic to a non-relativistic regime, introducing a new characteristic timescale in the thermal history. Together, these effects lead to simple analytical expressions for the reheating temperature. When the intermediate scalar is identified with the field responsible for the spontaneous breaking of lepton number, the same framework naturally connects the reheating temperature to the origin of neutrino masses and provides a well-motivated setting for sterile-neutrino dark matter.
Read the full abstract

Astrophysics of Galaxies

The Thesan-Zoom project: bursty star formation is incompatible with prolonged dust survival

Enrico Garaldi, Filip Popovic, Rahul Kannan et al. (+11 more)

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 $1500$ Å 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.
Read the full abstract

All authors Enrico Garaldi, Filip Popovic, Rahul Kannan, Aaron Smith, Ewald Puchwein, Naoki Yoshida, Kentaro Nagamine, Celine Peroux, Laura Keating, Mark Vogelsberger, William McClymont, Xuejian Shen, Sandro Tacchella, Lars Hernquist

High Energy Astrophysical Phenomena

Star-Disk Collisions II: Debris Stream Dynamics and Implications for QPEs and Other Transients Near SMBHs

Philippe Z. Yao, Eliot Quataert, Yan-Fei Jiang et al. (+1 more)

Quasi-periodic eruptions (QPEs) are repeating soft X-ray nuclear transients with recurrence times of hours-days and flare duty cycles of $\sim$10-20%. Many aspects of QPEs can be modeled as a stellar-mass orbiter that intersects an accretion disk producing a shocked debris cloud and a flare of radiation. We present three-dimensional Athena++ hydrodynamic simulations of star-disk interactions around a $10^{6}\,M_\odot$ supermassive black hole, including the black hole's tidal potential, the disk's Keplerian rotation, and orbital periods similar to those observed. After each disk encounter, freshly stripped stellar debris exits the Hill sphere to form an extended, asymmetric, roughly triaxial stream. Subsequent stream-disk collisions shock both stellar debris and disk gas to high specific energies and drive a wind-like outflow. At larger orbital periods the shocked stellar debris dominates the high specific energy debris, while at shorter orbital periods the shocked disk energy can be similar. From the shocked stellar mass measured in the simulations over time, we infer flare durations set by the time it takes the stellar debris stream to collide with the disk, consistent with the observed constant duty cycle of $\sim$10-20%, independent of orbital period. The total shocked debris energy is consistent with QPE flare energetics. Our results favor one observable flare per stellar orbit except perhaps at the shortest orbital periods where the shocked star and disk energetics can be similar. Variations in the stream's center of mass relative to the star, the stream density, and other properties can produce diverse changes in the time of the flare's peak relative to the time of the star-disk collision. We discuss the implications of our results for QPE timing and for other transients in galactic nuclei.
Read the full abstract

All authors Philippe Z. Yao, Eliot Quataert, Yan-Fei Jiang, Itai Linial

High Energy Astrophysical Phenomena

The type Ia supernova 2023vjh: a peculiar 1991bg-like SN with unusually faint light curves

M. Kopsacheili, L. Galbany, G. Folatelli et al. (+32 more)

We present observations of the 1991bg-like type Ia supernova (SN) 2023vjh, associated with the elliptical galaxy MCG+04-10-013. Its light-curve shape parameters ($\Delta m_{15}(B)=1.89 \pm 0.01$ mag and $s_{BV} = 0.45 \pm 0.03$), together with its spectroscopic evolution, place it firmly within the class of fast-declining, subluminous SNe Ia. The near-peak spectra show prominent features of Si II, Ca II, and Ti II, consistent with a cool photosphere, and place SN 2023vjh within the "cool" and extreme cool regions on the classification diagrams. In addition, the three late-phase near-infrared (NIR) spectra display the Ca II NIR triplet, Fe II, and Co II absorptions, but no obvious $H$-band break. Although SN 2023vjh falls in the same regions of the classification diagrams as other well-studied 91bg-like events, it shows some deviation within this class. In particular, it is systematically fainter than predicted by explosion models and fainter than other well-studied 91bg-like SNe. Light-curve fitting and color-based analyses indicate a relatively large reddening ($E(B-V)_{host}\sim$ 0.2 - 0.35 mag), which is unusual for 91bg-like SNe. Yet, the lack of detectable Na I D absorption in its spectra, along with its large projected distance from the center of its passive host galaxy (6.8 kpc), suggests that interstellar extinction along the line of sight is minimal. SN 2023vjh appears to be fainter than typical 91bg-like SNe, and it could be affected by circumstellar material (CSM). Comparisons with explosion models and alternative extinction prescriptions show that including CSM-like extinction improves agreement in the blue bands, but residual discrepancies in the $i$-band may reflect limitations in the model.
Read the full abstract

All authors M. Kopsacheili, L. Galbany, G. Folatelli, M. M. Phillips, C. R. Burns, M. D. Stritzinger, H.-Y. Miao, M. González-Bañuelos, R. García-Benito, T. E. Müller-Bravo, E. Y. Hsiao, K. Auchettl, J. P. Anderson, C. Ashall, T.-W. Chen, D. D. Desai, M. E. Huber, T. de Jaeger, J. M. DerKacy, M. Gromadzki, J. T. Hinkle, W. Hoogendam, C. Jimenez-Palau, K. Matilaine, P. A. Mazzali, K. Medler, T. Pessi, C. Pfeffer, K. Phan, G. Pignata, R. Sanfeliu, B. J. Shappee, M. A. Tucker, H. Xiao, D. R. Young

Astrophysics of Galaxies

What becomes of JWST/NIRCam-selected high-redshift massive galaxies?

Devontae C. Baxter, Alison L. Coil, Ethan O Nadler et al. (+2 more)

Early JWST/NIRCam surveys revealed a puzzling population of high-redshift massive galaxy candidates largely absent from previous rest-frame UV surveys. Spectroscopic follow-up has both confirmed and contested these candidates, whose potential overabundance may challenge traditional models of galaxy formation. In this study we evaluate the reliability of the photometric selections used to identify these candidates in observational data by applying them to galaxies in the TNG300 simulation with synthetic dust-attenuated photometry. Among the five observational selection criteria considered, we find that the selection presented by Pérez-González et al. is the most reliable and inclusive. Nevertheless, only 1 of the 18 galaxies at $z\sim5$ with $M_{\star} \geq 10^{11}~\mathrm{M_{\odot}}$ in the simulation satisfies this selection; the remaining 17 galaxies are on average $\sim0.5$ mag bluer than the color selection under the adopted dust model. We construct an improved JWST/NIRCam color-magnitude selection that provides a more complete census of massive galaxies at $z\sim5$ in TNG300 while excluding dusty, low-mass galaxies identified by criteria from the observational literature. We investigate the descendants of the most massive NIRCam-selected galaxies at $z=7,~4,$ and $2$ in TNG300, finding that they rarely evolve into the most massive galaxies by $z=0$. In general only the high-redshift massive galaxies that undergo substantial late-time ($z\lesssim0.2$) merger-driven growth become the most massive galaxies in the Universe today. Together these results suggest that current observational JWST/NIRCam selections are not identifying the most massive high-redshift galaxies, and caution against interpreting high-redshift massive galaxies as direct progenitors of the most massive galaxies at $z=0$.
Read the full abstract

All authors Devontae C. Baxter, Alison L. Coil, Ethan O Nadler, Xuejian Shen, Mark Vogelsberger

Astrophysics of Galaxies

Dark matter density profiles of the Milky Way satellite population: reconciling simulations and observations

Jorge Sarrato-Alós and José María Arroyo-Polonio

The cusp-core problem remains a key challenge for the $\Lambda$CDM model. Historically, comparing inner dark matter halo slopes ($d\log\rho/d\log r$) from simulations and observations has suffered from a methodological mismatch: evaluating slopes at different radii creates apparent tensions due to inconsistent definitions rather than genuine physical differences. We rigorously compare dark matter density profiles of Milky Way (MW) dwarf satellites inferred from dynamical modelling against CDM hydrodynamical simulations. Using the NIHAO and FIRE-2 suites ($M_{\star} \sim 10^3$-$10^{11}\,\rm M_\odot$), we confront simulated profiles with dynamical inferences of 16 MW satellites across four methods, strictly within their reliability limits. Crucially, we evaluate simulated slopes at observationally accessible radii, directly comparing profiles at matched stellar mass. Both observed and simulated galaxies exhibit a mass-dependent core-formation trend: inner slopes rise from cuspy values ($\approx -1.5$) at $M_{\star} \sim 10^5\,\rm M_\odot$ to core-like values ($\approx 0$ to $-0.4$) at $M_{\star} \sim 10^8\,\rm M_\odot$. Efficient core formation begins at $M_{\star} \gtrsim 10^6\,\rm M_\odot$ for centrals, where satellites display increased scatter consistent with tidal disruption. Simulated profiles agree with observations within $\lesssim 1\sigma$ for most systems, outperforming a NFW profile. Furthermore, differing observational models present scatter for a given system comparable to the simulation-to-observation offset. Evaluating dark matter density slopes at fixed physical radii significantly improves agreement between hydrodynamical simulations and dynamical models, revealing no clear tension when comparing profiles at fixed stellar mass, but Willman\,1 remains an unresolved outlier.
Read the full abstract

Astrophysics of Galaxies

Magnetogenesis by galactic processes: impact on circumgalactic and intergalactic fields

Rahul Ramesh, Enrico Garaldi and Chris Byrohl

We investigate the origin and evolution of cosmic magnetic fields using a suite of large-volume cosmological magnetohydrodynamic simulations (L$_\mathrm{box}=25$ Mpc/h) run with the moving-mesh code AREPO. Atop the IllustrisTNG galaxy formation model, we implement additional recipes for magnetogenesis in which magnetic energy is injected during supernovae (SNe) and supermassive black hole (SMBH) feedback events, and compare these to simulations initialized with uniform primordial seed fields. Halo magnetic field strengths at $z=0$ are largely similar across seeding models and are primarily amplified and sustained by small-scale and halo-scale dynamo action. Nevertheless, we find differences in magnetic field topology, with SMBH-driven models exhibiting systematically smaller coherence lengths than primordial-only and SNe-only runs. We find that feedback-driven injection accelerates the onset of dynamo growth, leading to more rapid convergence of magnetic field strengths with numerical resolution, particularly in low-mass halos. In the intergalactic medium (IGM), SNe-only injection underproduces magnetic fields relative to inferred lower limits from $\gamma$-ray cascade constraints at both $z=0$ and $z \sim 3$, whereas our specific SMBH-based injection prescription satisfies present-day constraints but remains in mild tension at high redshifts. Reconciling these specific high-$z$ constraints therefore likely requires either modified feedback prescriptions or an additional primordial seeding component.
Read the full abstract

Instrumentation and Methods for Astrophysics

DegenDetector: Symbolic Recovery of Parameter Degeneracies in Bayesian Posteriors

Chaipat Tirapongprasert and Matthew Ho

We introduce DegenDetector, a framework for identifying and characterizing parameter degeneracies in posterior distributions as closed-form symbolic equations. By combining mutual information screening with alternating symbolic regression, we facilitate automated and interpretable identification of degenerate relationships without domain-specific input. While standard tools such as corner plots can indicate that correlations exist, they do not reveal the underlying functional form. DegenDetector fills this gap by expressing multi-parameter degeneracies as closed-form equations, providing interpretable structure that scales to high-order parameter spaces.
Read the full abstract

Earth and Planetary Astrophysics

Debris Disc Substructures Induced by Secular Planetary Perturbations

Roman R. Rafikov, Deniz Akansoy and Antranik A. Sefilian

Observations of debris discs have the potential to provide valuable information about massive planets perturbing them gravitationally. In this work, we explore the evolution of the azimuthally-averaged (or axisymmetric) surface density (ASD) — a characteristic routinely derived from observations — in a disc that is secularly perturbed by an inner planet. We develop a detailed analytical understanding of ASD evolution and verify it using a novel numerical framework debrispy, which we make publicly available. With these tools we show that in a secularly evolving disc ASD develops a set of sharp features — sharp peaks at eccentricity nulls and finite discontinuities at caustic points where particle periastra or apoastra pile up — marching out through the disc as it ages. We probe the dependence of these features on planetary eccentricity, ratio of the free to forced particle eccentricity, and underlying radial mass distribution, showing in particular that more eccentric planets produce more prominent ASD features. Convolution with the PSF of realistic observations (as well as the non-zero random free eccentricity of debris) smooths out these features, but they can still be detectable in high-resolution observations. The radial locations of secular ASD peaks follow a particular, well-defined pattern, which should unambiguously point to their secular nature in observations. We illustrate how both detection and non-detection of such secular features in observed discs can be used (in combination with other constraints) to measure or constrain key parameters of perturbing planets (even those not yet detected) — mass, semi-major axis and eccentricity.
Read the full abstract

Astrophysics of Galaxies

Deep Spectroscopic Follow-Up of Maisie's Galaxy – A Typical Galaxy in the Early Universe

Rebecca L. Larson, Taylor A. Hutchison, Steven L. Finkelstein et al. (+26 more)

The first several years of JWST observations have yielded surprisingly large numbers of bright $z>10$ galaxies, with follow-up spectroscopy of many of these sources implying extreme star formation activity and/or AGN content. Here, we present a combination of two deep Cycle 3 NIRSpec G395M programs, totaling over 19 hours of exposure time, plus MIRI/LRS observations for one such high-redshift source: Maisie's Galaxy. We provide an updated redshift measurement of $z = 11.408 \pm 0.005$ for this source. Measurements of the [OII] doublet in these data yield an electron density ($n_e = 108.56^{+873.9}_{-35.37}$) and a star-formation rate (SFR$_{[OII]} = 1.3 \pm 0.35$), placing it along the star-formation main sequence (SFMS) and indicating that this is a much more typical, rather than extreme, source in the early Universe. We also report fluxes for the [OIII]$\lambda$5008 and [NeIII]$\lambda$3869 lines that provide us with a $\log$(Ne3O2) $= -0.219 \pm 0.145$ and a $\log$(O32) $=0.724 \pm 0.191$. We estimate the metallicity ($Z/Z_{\odot} = 0.17 \pm 0.05$) and ionization parameter ($\log$(U) $= -2.26 \pm 0.13$) from the Ne3O2 ratio. We place this galaxy in the context of other $z>10$ sources with similar line detections and compare the results to those obtained from SED fitting. The results suggest that we should go deeper with our observations to better understand the average galaxy population at these early times.
Read the full abstract

All authors Rebecca L. Larson, Taylor A. Hutchison, Steven L. Finkelstein, Pablo Arrabal Haro, Casey Papovich, Weida Hu, Javier Álvarez-Márquez, Ruqiu Lin, Jorge A. Zavala, Volker Bromm, Nikko J. Cleri, Abdurro'uf, Brittany Vanderhoof, Bren E. Backhaus, Dan Coe, Henry C. Ferguson, Ananya Ganapathy, Norman A. Grogin, Michaela Hirschmann, Intae Jung, Jeyhan S. Kartaltepe, Anton M. Koekemoer, Dale D. Kocevski, Ray A. Lucas, Alexa M. Morales, Pablo G. Pérez-González, Nor Pirzkal, Jonathan R. Trump, L. Y. Aaron Yung

High Energy Astrophysical Phenomena · Cross-listed from hep-th

Decoupling Limit of Quiver Theories and the Angular Spectra of Extreme C-metrics

Peng Yang and Kilar Zhang

We investigate the angular eigenvalue problem of the extreme charged C-metric. Starting from a constrained five-point Fuchsian description of the non-extremal angular equation, we obtain its extreme limit by decoupling a hypermultiplet in the dual four-dimensional $\mathcal{N}=2$, $\mathrm{SU(2)}\times \mathrm{SU(2)}$ linear quiver gauge theory. We derive the parameter dictionary and renormalized Matone relations, and connect the physical limit $Q\to M$ to the quiver decoupling by an explicit coordinate transformation. Regular endpoint conditions give an algebraic quantization condition. Eigenvalues obtained from the truncated instanton prepotential are tested against independent angular spectra near extremality and at larger acceleration. The axisymmetric sector recovers the known exact spectrum.
Read the full abstract

High Energy Astrophysical Phenomena

SACRA-K: A Performance-Portable Numerical Relativity Code with Kokkos

Ming-Zhe Han, Kenta Kiuchi and Masaru Shibata

We present SACRA-K (SimulAtor for Compact objects in Relativistic Astrophysics with Kokkos), a performance-portable numerical relativity code ported from the Fortran code SACRA-MPI to C++ with the Kokkos library. SACRA-K retains the physics and numerical methods of SACRA-MPI, namely a Baumgarte-Shapiro-Shibata-Nakamura (BSSN) spacetime evolution with Z4c constraint propagation and a box-in-box Berger-Oliger adaptive mesh refinement, together with the high resolution shock capturing scheme for the general relativistic hydrodynamics from NANASI, while gaining cross vendor portability. We validate the port against SACRA-MPI across binary black hole, black hole neutron star, and binary neutron star configurations. Across these tests, the waveform discrepancies are well below both the scatter among independent numerical relativity codes and the resolution dependent variation within a given code, and remain at or below the distinguishability threshold of current gravitational-wave detectors; the $\pi$ symmetry is preserved exactly at the bitwise level; and the gravitational wave phase of the binary neutron star merger exhibits second order convergence. In the smallest test configuration, SACRA-K runs about an order of magnitude faster on the NVIDIA A100 GPU cluster or the AMD MI300A APU cluster than the Fortran SACRA-MPI on the CPU cluster, and we measure its scaling up to 256 accelerator devices.
Read the full abstract