https://arxiv.org/api/LZ1SXrkY/tbkv3uJ7c+1Pa4G4Wg 2026-06-09T20:35:21Z 35880 0 15 http://arxiv.org/abs/2605.30412v2 2026-06-08T17:37:09Z

We present a set of metrics and methods for testing and comparing a range of modern gradient-free Markov Chain Monte Carlo (MCMC) samplers against the commonly used Metropolis-Hastings (MH) algorithm. The goal is to quantify key performance metrics, including sampler ergodicity, robustness and overall likelihood performance. To provide a controlled and interpretable testbed, we use the Rosenbrock function and Neal's funnel as representative unimodal cases, while Gaussian random likelihood landscapes in three, five, and eight dimensions serve as multimodal test scenarios. The samplers considered include affine-invariant moves from the literature, such as the stretch and walk moves, the differential evolution move, and the snooker move. We additionally introduce two novel variations: a modified stretch move that incorporates a Principal Component Analysis (PCA) transformation, and a hybrid blend move that combines features of both differential evolution and stretch dynamics. Beyond sampler evaluation, we demonstrate reconstructing likelihood landscapes from sampled points using a quadtree algorithm. Additionally, we explore the use of optimisation algorithms to refine the best parameter set in terms of its likelihood, and find consistent improvements in log-likelihood values, with the post-sampling gain becoming more significant in higher-dimensional problems. Our comparative results of sampler testing show that the differential evolution algorithm, when tuned to a target acceptance fraction of 25%, consistently outperforms all other samplers in terms of ergodicity, robustness, and likelihood performance.

2026-05-28T18:00:00Z Submitted to MNRAS; 30 pages, 29 figures; resubmitted to fix bibliography Colin M. Poppelaars Marcel P. van Daalen http://arxiv.org/abs/2606.07256v2 2026-06-08T17:23:40Z

The development of modern high-energy astrophysics has been deeply intertwined with advances in particle detector technology. Guido Barbiellini Amidei (1943-2024) played a pivotal role in bridging experimental particle physics and astrophysical observation. His scientific career spanned over four decades, from early electron-positron collider experiments at ADONE and LEP (DELPHI) to space based missions such as AGILE, Fermi, and PAMELA. This memorial paper reviews the evolution of high-energy astrophysics as a detector-driven science, highlighting key domains where Barbiellini left an indelible mark: gamma-ray astronomy, cosmic-ray physics, and antimatter studies. We discuss his personal contributions to silicon tracking, calorimetry, data analysis, and his leadership in international collaborations. The conceptual impact of his interdisciplinary approach is examined, and future perspectives in the observation of the high-energy universe are outlined, recognizing that the path forward is built on the foundations he helped lay

2026-06-05T13:30:26Z 8 pages. Published in Condensed Matter, special issue "The Universe Observed With Particle Detectors: Celebrating the Scientific Legacy of Prof. Guido Barbiellini Amidei" Roberto Capuzzo Dolcetta Dep. of Physics, Sapienza, University of Roma, Piazzale A. Moro 2, 00185 Rome, Italy Istituto Nazionale di Fisica Nucleare Centro di Ricerche E. Fermi, via Panisperna 89, 00184 Rome, Italy Istituto Nazionale di Astrofisica, viale del Parco Mellini 8, 00136 Rome, Italy 10.3390/condmat11020022 http://arxiv.org/abs/2606.09713v1 2026-06-08T16:38:36Z

Domain walls provide an excellent fit to Pulsar Timing Array (PTA) data. A distinctive feature of the associated gravitational-wave (GW) spectrum is its ultraviolet (UV) tail, which can extend into the mHz band and thereby make cross-detection with space-based interferometers such as LISA and Taiji possible. In this work, we explore the PTA-compatible parameter space of domain-wall models and study both the reconstruction prospects at LISA and Taiji and the information gain achievable through joint PTA--LISA and PTA--Taiji analyses. We find that LISA and Taiji can probe an extended region of parameter space yielding an ultraviolet tail in the milli-Hz band, even in the presence of astrophysical foregrounds. Within the PTA-supported region, however, the genuinely informative regime is more restricted and concentrated toward the high-signal edge. In the space-based-detector-only analysis, the posterior is strongly degenerate in the underlying model parameters, although one principal parameter combination can be reconstructed with high precision where the signal is sufficiently strong. When PTA-informed priors are incorporated, the additional information gain is localized to the same high-signal region: there the space-based data produce a non-negligible posterior update and strongly suppress the catastrophic LISA(Taiji)-only degeneracy, while the final posterior often retains an axis ratio comparable to that of PTA alone. Our analysis is based on a 10-dimensional Bayesian inference with two domain-wall signal parameters and eight nuisance parameters describing instrumental noise and astrophysical foregrounds. We use 49 grid injections for the LISA-only and Taiji-only analyses and 65 injections from the PTA-supported region for the joint analyses, with Clough--Tocher interpolation used to construct posterior heat maps.

2026-06-08T16:38:36Z 35 pages, 10 figures Satyabrata Datta Rome Samanta http://arxiv.org/abs/2512.15669v5 2026-06-08T16:15:58Z

This work aims to choose potential astronomical sites that can be candidates for a new astronomical optical observatory in Ethiopia, in addition to the Entoto Observatory and Lalibela sites. For our primary investigation, the six basic criteria, namely the altitude of the mountains, artificial light pollution, cloud coverage, humidity, wind speed, and wind direction, were taken into account. Consequently, using the multi-criteria statistical Decision analysis (MCDSA) techniques, 21 high-potential places are selected and presented for further investigation out of 367 mountains. Among these 21 selected places, three sites, Bauhit, Meseraia, and T'at'a are the most suitable places for optical astronomy in Ethiopia. Those selected mountains are mapped and presented to study the future of the astronomical seeing effect. This study may contribute to the protection of those potential astronomical sites and their dark skies and the development of astrotourism for the sustainable development of modern astronomy in Ethiopia and in the East African region.

2025-12-17T18:22:39Z 5 pages, 3 tables and two figures SSGI Collbaration : Shambel Sahlu Nebiyu Suleyman Gemechu M. Kumssa Solomon T. Belay Feraol F. Dirirsa Esubalew Mulugeta Melesse Getnet Mirjana Povic Seblu H. Negu Betelehem Bilata Jerusalem Tamrat http://arxiv.org/abs/2606.09665v1 2026-06-08T15:48:37Z

The Vera C. Rubin Observatory Legacy Survey of Space and Time (LSST) is expected to observe active galactic nuclei (AGN) at sky densities of approximately 1000-4000 per sq. deg, enabling photometric reverberation mapping on an unprecedented scale. We present a meta-learning framework for AGN photometric reverberation mapping based on Attentive Latent Neural Processes (ALNP), developed by the SER-SAG-S1 directable software in-kind team for LSST. The framework clusters AGN light curves with similar topologies using Self-Organizing Maps and combines ALNPs with Mixture Density Models to learn light-curve structure, supermassive black hole (SMBH) properties, and accretion-disk transfer functions in an unsupervised manner. We evaluate the framework on simulated AGN light curves spanning a range of cadences and transfer functions, as well as on real data from the Zwicky Transient Facility. The learned latent representations encode information on both transfer functions and SMBH parameters. Relative to ensemble-trained baseline regressors, including Gaussian-process models, the framework improves light-curve reconstruction by 60-70%. The transfer function recovery improves by approximately 35% relative to the training prior in a low-variability cluster, while recovery of intrinsic SMBH and red-noise parameters improves by approximately 34%. We further demonstrate that models trained on simulated data can be applied to real AGN light curves. These results indicate that ALNP-based representations provide a flexible and scalable approach to photometric reverberation mapping and are well suited to the diverse AGN population expected from LSST and future time-domain surveys.

2026-06-08T15:48:37Z 34 pages, 25 figures, 13 tables, accepted in A&A (June 1st, 2026) Aman N. Raju Andjelka B. Kovačević Dragana Ilić Francesco Tombesi Luka Č. Popović Eric Slezak Paula Sanchez-Saez Marina Pavlović Iva Čvorović-Hajdinjak Saša Simić Đorđe Savić http://arxiv.org/abs/2605.30249v2 2026-06-08T15:35:01Z

The most irreplaceable capability of HST in the 2030s is not only its angular resolution or its UV--optical sensitivity, but its accumulated time baseline. We recommend a Hubble Local Group Astrometric Legacy that would obtain matched 2030s ACS/WFC and WFC3/UVIS imaging of selected M31 and M33 star cluster fields, combine those observations with a careful astrometric audit of the existing archive, and deliver public transverse velocity products through MAST. PHAT and PHATTER already provide exceptional first epochs across the nearest large external spirals, resolving tens to hundreds of millions of stars and identifying thousands of star clusters and compact background reference objects. Earlier targeted HST programs, dating back to the mid 90's, can extend the temporal baseline in selected cases, but only after camera, chip, filter, dithering, crowding, and reference-frame triage. A coherent 2030s repeat campaign would turn the best of this archive into bulk cluster motions, enabling the first systematic orbital mapping of star cluster populations beyond the Milky Way. These measurements would constrain disk heating, cluster disruption, accretion histories, stream associations, the M31--M33 interaction, and the mass distributions of Local Group spirals. The same program would also preserve and stress-test the calibration infrastructure needed for high-stability optical/UV astrophysics in the HWO era: geometric-distortion solutions, PSF and CTE modeling, matched-epoch observing strategies, long-term reference frames, and durable public high-level products.

2026-05-28T17:13:26Z 6 pages, 1 figure; white paper submitted to the STScI call for Hubble science into the 2030s; references corrected Borja Anguiano CEFCA Benjamin J. Gibson STScI Sten Hasselquist STScI http://arxiv.org/abs/2606.09576v1 2026-06-08T14:51:52Z

Stellar streams are thin, elongated collections of stars formed by gravitational disruption of orbiting star clusters or dwarf galaxies and are highly sensitive probes of the Milky Way's dark matter distribution and formation history. We present $\texttt{SCREAM}$ ($\textbf{S}$tream $\textbf{C}$ha$\textbf{R}$acterization with $\textbf{E}$rror $\textbf{A}$ware $\textbf{M}$achine Learning), a weakly-supervised framework to identify member stars of stellar streams. Building on the $\texttt{CATHODE}$ method originally developed for particle physics, $\texttt{SCREAM}$ identifies streams as localized feature-space over-densities, avoiding rigid physical priors like assumed gravitational potentials or strict isochrone filtering. Crucially, $\texttt{SCREAM}$ is the first machine learning (ML) framework in this domain to directly incorporate observational uncertainties into the neural network training objective. Using astrometric and photometric data from Gaia Data Release 3 and the Dark Energy Spectroscopic Instrument (DESI) Legacy imaging survey, we demonstrate our algorithm's performance on the prominent GD-1 stream. Validated against independent labels, $\texttt{SCREAM}$ achieves an F1 score of 0.745, substantially outperforming existing ML methods in both precision and recall. Furthermore, $\texttt{SCREAM}$ recovers the physically expected diffuse "cocoon" of GD-1 and faint main-sequence members that classical physics-based algorithms (e.g., $\texttt{STREAMFINDER}$) miss. Our results highlight the transformative potential of uncertainty-aware, weakly-supervised ML to uncover complex galactic structures.

2026-06-08T14:51:52Z Accepted for Conference on Physics and AI at Stanford University (PAI2026) Alexandros Pratsos Biprateep Dey Ting S. Li http://arxiv.org/abs/2606.04875v2 2026-06-08T14:13:21Z

Multidimensional Gaussian Process (multi-GP) regression is widely used to disentangle stellar and planetary signals in radial velocities (RVs) by jointly modelling ancillary activity indicators. However, identifying the combination of indicators that best constrains the stellar signal in the RVs is non-trivial, as classical model comparison methods are not directly applicable when multi-GPs involve different time series combinations. In this work, we present an information criterion to compare multi-GP models based on their ability to explain the RV component, $\mathrm{MGIC}_{\rm rv}$. This metric combines the conditional RV likelihood with an effective parameter count that accounts for the regularisation imposed by the multi-GP model on the RV component. We demonstrate that $\mathrm{MGIC}_{\rm rv}$ provides a quantitative and robust framework for multi-GP model comparison, identifying the activity indicators that most effectively constrain the RV signal. Although developed in the context of RV analysis, the proposed criterion is general and applicable to multi-GP problems in which the inference focuses on a specific observable.

2026-06-03T13:38:31Z Accepted for publication in MNRAS letters Oscar Barragán 10.1093/mnras/stag1054 http://arxiv.org/abs/2606.09515v1 2026-06-08T14:06:37Z

Millimeter-wave Adaptive Optics (MAO) is essential for high-precision large-aperture submillimeter telescopes, requiring real-time compensation of wavefront errors by capturing them as spatially-discrete excess path length (EPL) fluctuations. This paper presents a unified control-theoretic framework for the EPL compensation problem. We first model the optical drive system as a plant where input commands relate to measured EPL through a first-order system representing mechanical response delay and a measurement matrix characterizing the actuator-to-sensor coupling. We mathematically formulate the control task as an asymptotic disturbance suppression problem, specifically targeting low-frequency disturbances such as thermal and wind-induced deformations. Second, we propose an Anti-Windup Proportional-Integral (AWPI) control law. By employing a decoupling strategy, the design is reduced to a loop-shaping problem for decoupled scalar sensitivity functions, ensuring both stability margins and asymptotic disturbance suppression of constant-valued disturbances. The anti-windup mechanism is integrated to maintain control continuity during the recovery from saturation, preventing undesirable discontinuities in the drive command. Third, we introduce practical operational tools: a manual focus adjustment scheme that allows observer intervention without interfering with the feedback loop, and the cosine similarity index to quantify the suppressibility of specific Zernike modes. Numerical simulations, incorporating a three-axis secondary reflector drive and five-point EPL measurements, demonstrate direction-dependent disturbance rejection and the suppression of von Karman-modeled wind turbulence, validating the effectiveness of the proposed framework for real-world telescope applications.

2026-06-08T14:06:37Z 44 pages, 20 figures. Accepted for publication in JATIS Ichiro Jikuya Yoichi Tamura http://arxiv.org/abs/2507.18683v2 2026-06-08T13:08:43Z

Understanding the structure of our universe and the distribution of matter is an area of active research. As cosmological surveys grow in complexity, the development of emulators to efficiently and effectively predict matter power spectra is essential. We are particularly motivated by the Mira-Titan Universe simulation suite that, for a specified cosmological parameterization (termed a "cosmology"), provides multiple response curves of various fidelities, including correlated functional realizations. Our objective is two-fold. First, we estimate the underlying matter power spectra, with appropriate uncertainty quantification (UQ), from all of the provided curves. To this end, we propose a novel Bayesian deep Gaussian process (DGP) hierarchical model which synthesizes all the simulation information to estimate the underlying matter power spectra while providing effective UQ. Our model extends previous work on Bayesian DGPs from scalar responses to correlated functional outputs. Second, we leverage our predicted power spectra from various cosmologies in order to accurately predict the entire matter power spectra for an unobserved cosmology. For this task, we use basis function representations of the functional spectra to train a separate Gaussian process emulator. Our method performs well in synthetic exercises and against the benchmark cosmological emulator (Cosmic Emu).

2025-07-24T16:48:38Z 22 pages, 14 figures. Revised and accepted version for publication in Data Science in Science Stephen A. Walsh Annie S. Booth David Higdon Jared Clark Kelly R. Moran Katrin Heitmann http://arxiv.org/abs/2606.02778v2 2026-06-08T12:39:10Z

I present EXOVEIL, a transit detection system that learns what a star's brightness should look like and flags when reality disagrees. Unlike existing systems that require phase-folded input, EXOVEIL operates on raw flux time series and can detect planets that transit only once.A Transformer world model, trained on 16,499 Kepler light curves with transit-masked self-supervised learning, predicts expected stellar flux. A matched-filter detector with variance weighting extracts transit signals from the prediction residuals. A learned classifier (XGBoost) separates planets from false positives, achieving AUC 0.938 on Kepler DR25. Applied to single-transit injection-recovery, EXOVEIL recovers 32% of transits at 1000 ppm depth a task where all classification-based systems score 0% by construction. A blind search of 3,737 Kepler stars yields 179 new transit-like signals not present in the DR25 TCE catalogue, including 46 monotransit candidates. Applied withoutretraining to 47 confirmed TESS planets in the PLATO LOPS2 field, EXOVEIL achieves 100% recovery, demonstrating zero-shot cross-mission transfer. At PLATO's 25-second cadence, detection reaches 100 ppm -- approaching the Earth-analog regime. I provide the first application of conformal prediction to transit detection (95.9% empirical coverage) and release the system as pip install exoveil with pretrained weights and a candidate catalogue.

2026-06-01T18:41:10Z v2: Adds gap-proximity vetting (45% of candidates flagged as near-gap), head-to-head TLS comparison in monotransit mode, and new headline candidate KIC 12253350 replacing KIC 11706231. ~9 pages, 6 figures, 4 tables. pip install exoveil (v0.2.0) Pratik Priyanshu http://arxiv.org/abs/2606.09309v1 2026-06-08T10:16:02Z

Fast, highly sensitive, and broadband detectors operating in the mid-to-far-infrared (MIR-FIR) spectral region are essential for applications ranging from astrophysics to time-resolved spectroscopy using laboratory-based sources and beamlines at Infrared Free-Electron Laser (IR-FEL) facilities. Conventional Lumped Element Kinetic Inductance Detectors (LEKID) achieve high optical efficiency using resonant quarter wavelength (λ/4) backshort cavities. However, this cavity-based approach is inherently narrowband and becomes optomechanically challenging at the mid to far infrared (MIR-FIR) wavelengths. Here, we present a completely backshort-free LEKID absorber architecture that achieves broadband absorption exceeding 50%. The meander absorbers were fabricated from a superconducting Ti-V alloy and characterized using IR radiation from the IR-FEL at RRCAT, India. By illuminating through the silicon substrate and matching the meander's sheet resistance to the silicon substrate's wave impedance, the front-side reflection is strongly suppressed. Simultaneously, the sub-wavelength periodicity of the dense meander inhibits propagating transmission into free space via evanescent wave confinement. This combined mechanism drives efficient Ohmic dissipation within the metallic meander, yielding an experimental absorption efficiency ranging from 50% to 90% across the 12.5-30 micrometer wavelength range. The cavity-free design greatly simplifies the fabrication of the focal plane array while providing the broadband response required for next-generation detectors. To assess the suitability of the Ti40V60 alloy as an active superconducting detector material, a test LEKID resonator was also designed, fabricated, and experimentally characterized.

2026-06-08T10:16:02Z 22 Pages, 7 Figures Shilpam Sharma Shekhar Chandra Pandey Anudeep Singh Shankar Lal R. S. Saini M. K. Chattopadhyay http://arxiv.org/abs/2601.23106v3 2026-06-08T10:13:37Z

The next-generation of cryogenic neutrinoless double-beta decay experiments require increasingly fast readout in order to improve background discrimination. These experiments, operated as cryogenic calorimeters at $\sim$10 mK, are usually read out by high-impedance neutron transmutation doped (NTD) thermistors, which provide good energy resolution, but are limited by $\sim$1 ms response times. Superconducting detectors, such as transition-edge sensors (TESs) with a time resolution of $\sim$100 $μ$s, offer superior timing performance over NTD semiconductor bolometers. To make this technology viable for an application to a thousand or more channels, multiplexed readout is necessary in order to minimize the thermal load and radioactive contamination induced by the readout. Frequency-domain multiplexing readout (fMUX) for TESs, previously developed at Berkeley Lab and McGill University, is currently in use for mm-wave telescopes with detector sampling rates in the order of 100 Hz. We demonstrate a new readout system, based on the McGill/Berkeley digital fMux readout, to satisfy the higher bandwidth and noise requirements of the next generation of TES-instrumented cryogenic calorimeters. Each multiplexing readout module comprises 10 superconducting resonators in the 1--5 MHz range and a DC superconducting quantum interference device (DC-SQUID), interfaced to high-speed field programmable gate array (FPGA)-based electronics for digital signal processing and low-latency SQUID feedback. The new readout samples detectors at 156 kHz, three orders of magnitude faster than its cosmology-oriented predecessor, and demonstrates a stable feedback bandwidth of 3 kHz in a real TES-based system.

2026-01-30T15:52:03Z JINST 21 (2026) 06, P06007 M. Adamič McGill & LBNL M. Beretta UCB & INFN J. Camilleri LBNL & Virginia Tech C. Capelli LBNL & Zurich U. M. A. Dobbs McGill T. Elleflot LBNL B. K. Fujikawa LBNL Yu. G. Kolomensky LBNL & UCB D. Mayer MIT J. Montgomery McGill V. Novosad ANL A. M. Sindhwad UCB V. Singh UCB G. Smecher t0.technology A. Suzuki LBNL B. Welliver UCB 10.1088/1748-0221/21/06/P06007 http://arxiv.org/abs/2606.09238v1 2026-06-08T09:14:53Z

Extreme mass ratio inspirals (EMRIs) are among the primary targets for space-based gravitational-wave (GW) detectors, providing valuable opportunities to study stellar-mass compact objects orbiting supermassive black holes (SMBHs) and to probe gravity in the strong-field regime. As the LISA, TAIJI, and TianQin missions are expected to operate around 2035, joint observations by multiple detectors may provide enhanced measurement capabilities compared to individual missions. In this work, we investigate the detectability and parameter-constraint prospects for EMRIs using global detector networks composed of LISA, TianQin, and two alternative TAIJI orbital configurations. Employing Fisher information matrix analysis, we find that joint observations improve source localization relative to a standalone LISA mission, with sky-localization uncertainties reduced by up to two orders of magnitude for a one-month observation period. We further find that a three-detector network (LISA-TAIJI-TianQin) operating for one month yields parameter constraints comparable to, and in some cases tighter than, those obtained from a one-year observation by LISA alone. This result indicates that network observations can partially compensate for shorter observation durations in parameter inference. Furthermore, we evaluate measurement uncertainties in concurrent dual-signal scenarios: (i) two EMRIs orbiting the same SMBH and (ii) two EMRIs with identical intrinsic parameters but different sky locations and orientations. The results indicate that differences in the detector responses associated with the source geometries reduce correlations between overlapping signals, allowing the sources to remain distinguishable even for a standalone mission. The inclusion of multiple detector constellations further improves source localization and tightens parameter constraints in these concurrent-source scenarios.

2026-06-08T09:14:53Z 22 pages, 7 figures, 4 tables; comments are welcome Chao Zhang Gang Wang http://arxiv.org/abs/2606.09219v1 2026-06-08T08:52:17Z

Source detection in modern observational astronomy is a cornerstone for localizing and identifying stellar sources accurately. It is crucial for studies such as stellar population synthesis and cosmological parameter estimation. However, the characteristics of astronomical images, including high density, the effect of point spread functions and low signal-to-noise ratios, significantly challenge the latest advanced object detectors. Besides, fully-supervised detection methods are hardly practical, due to the significant difficulty in annotating dense, small, and faint sources in astronomical images. To tackle the scarcity of astronomical datasets, we introduce a new comprehensive benchmark (LAMOST-DET), comprising 18,400 astronomical images and 728,898 source instances. Upon the dataset, we further devise a novel semi-supervised learning framework coined Nova Teacher, capable of detecting dense sources effectively given sparse annotations. It integrates source light enhancement module, confidence-guided pseudo-supervision, and cross-view complementary mining in a dual-teacher paradigm. Extensive experiments on LAMOST-DET show that, Nova Teacher consistently improves previous competitors by 4.04% and 5.22% mAP under two semi-supervised settings. Additionally, our method competes against other detectors on a natural image dataset, validating its generalization ability to various scenarios. The source code is available at https://github.com/AcWiz/NovaTeacher.

2026-06-08T08:52:17Z Longhan Feng Zihuang Cao Ali Luo Yuanhao Guo Shuilian Yao Yixin Guo Qi Jia Yu Liu