More Like this

1. LOTUS: A (Non-) LTE Optimization Tool for Uniform Derivation of Stellar Atmospheric Parameters

   https://doi.org/10.3847/1538-3881/acb7f0  

   Li, Yangyang; Ezzeddine, Rana (March 2023, The Astronomical Journal)

   Abstract Precise fundamental atmospheric stellar parameters and abundance determination of individual elements in stars are important for all stellar population studies. Non–local thermodynamic equilibrium (non-LTE; hereafter NLTE) models are often important for such high precision, however, can be computationally complex and expensive, which renders the models less utilized in spectroscopic analyses. To alleviate the computational burden of such models, we developed a robust 1D, NLTE fundamental atmospheric stellar parameter derivation tool, LOTUS , to determine the effective temperature T eff , surface gravity log g , metallicity [Fe/H], and microturbulent velocity v mic for FGK-type stars, from equivalent width (EW) measurements of Fe i and Fe ii lines. We utilize a generalized curve of growth method to take into account the EW dependencies of each Fe i and Fe ii line on the corresponding atmospheric stellar parameters. A global differential evolution optimization algorithm is then used to derive the fundamental parameters. Additionally, LOTUS can determine precise uncertainties for each stellar parameter using a Markov Chain Monte Carlo algorithm. We test and apply LOTUS on a sample of benchmark stars, as well as stars with available asteroseismic surface gravities from the K2 survey, and metal-poor stars from the Gaia-ESO and R -Process Alliance surveys. We find very good agreement between our NLTE-derived parameters in LOTUS to nonspectroscopic values on average within T eff = ±30 K, and log g = ±0.10 dex for benchmark stars. We provide open access of our code, as well as of the interpolated precomputed NLTE EW grids available on Github (the software is available on GitHub 3 3 https://github.com/Li-Yangyang/LOTUS under an MIT License, and version 0.1.1 (as the persistent version) is archived in Zenodo) and documentation with working examples on the Readthedocs book. 

   more » « less
2. Non-local thermodynamic equilibrium analysis of the methylidyne radical molecular lines in metal-poor stellar atmospheres

   https://doi.org/10.1051/0004-6361/202245503  

   Popa, S. A.; Hoppe, R.; Bergemann, M.; Hansen, C. J.; Plez, B.; Beers, T. C. (February 2023, Astronomy & Astrophysics)

   Aims. An analysis of the methylidyne (CH) radical in non-local thermodynamic equilibrium (NLTE) is performed for the physical conditions of cool stellar atmospheres typical of red giants (log ɡ = 2.0, T eff = 4500 K) and the Sun. The aim of the present work is to explore whether the G band of the CH molecule, which is commonly used in abundance diagnostics of carbon-enhanced metal-poor stars, is sensitive to NLTE effects. Methods. LTE and NLTE theoretical spectra were computed with the MULTI code. We used one-dimensional (1D) LTE hydrostatic MARCS model atmospheres with parameters representing eleven red giant stars with metallicities ranging from [Fe/H] = −4.0 to [Fe/H] = 0.0 and carbon-to-iron ratios of [C/Fe] = 0.0, +0.7, +1.5, and +3.0. The CH molecule model was represented by 1981 energy levels, 18 377 radiative bound-bound transitions, and 932 photo-dissociation reactions. The rates due to transitions caused by collisions with free electrons and hydrogen atoms were computed using classical recipes. Results. Our calculations suggest that NLTE effects in the statistical equilibrium of the CH molecule are significant and cannot be neglected for precision spectroscopic analysis of C abundances. The NLTE effects are mostly driven by radiative over-dissociation, owing to the very low dissociation threshold of the molecule and significant resonances in the photo-dissociation cross-sections. The NLTE effects in the G band increase with decreasing metallicity. When comparing the C abundances determined from the CH G band in LTE and in NLTE, we show that the C abundances are always under-estimated if LTE is assumed. The NLTE corrections to C abundance inferred from the CH feature range from +0.04 dex for the Sun to +0.21 dex for a red giant with metallicity [Fe/H] = −4.0. Conclusions. Departures from the LTE assumption in the CH molecule are non-negligible, and NLTE effects have to be taken into account in the diagnostic spectroscopy based on the CH lines. We show here that the NLTE effects in the optical CH lines are non-negligible for the Sun and red giant stars, but further calculations are warranted to investigate the effects in other regimes of stellar parameters. 

   more » « less
3. Fundamental stellar parameters of benchmark stars from CHARA interferometry: II. Dwarf stars

   https://doi.org/10.1051/0004-6361/202141833  

   Karovicova, I.; White, T. R.; Nordlander, T.; Casagrande, L.; Ireland, M.; Huber, D. (February 2022, Astronomy & Astrophysics)

   Context. Stellar models applied to large stellar surveys of the Milky Way need to be properly tested against a sample of stars with highly reliable fundamental stellar parameters. We have established a programme aiming to deliver such a sample of stars. Aims. Here we present new fundamental stellar parameters of nine dwarf stars that will be used as benchmark stars for large stellar surveys. One of these stars is the solar-twin 18 Sco, which is also one of the Gaia -ESO benchmarks. The goal is to reach a precision of 1% in effective temperature ( T eff ). This precision is important for accurate determinations of the full set of fundamental parameters and abundances of stars observed by the surveys. Methods. We observed HD 131156 ( ξ Boo), HD 146233 (18 Sco), HD 152391, HD 173701, HD 185395 ( θ Cyg), HD 186408 (16 Cyg A), HD 186427 (16 Cyg B), HD 190360, and HD 207978 (15 Peg) using the high angular resolution optical interferometric instrument PAVO at the CHARA Array. We derived limb-darkening corrections from 3D model atmospheres and determined T eff directly from the Stefan–Boltzmann relation, with an iterative procedure to interpolate over tables of bolometric corrections. Surface gravities were estimated from comparisons to Dartmouth stellar evolution model tracks. We collected spectroscopic observations from the ELODIE spectrograph and estimated metallicities ([Fe/H]) from a 1D non-local thermodynamic equilibrium (NLTE) abundance analysis of unblended lines of neutral and singly ionised iron. Results. For eight of the nine stars we measure the T eff ⪅ 1%, and for one star better than 2%. We determined the median uncertainties in log  g and [Fe/H] as 0.015 dex and 0.05 dex, respectively. Conclusions. This study presents updated fundamental stellar parameters of nine dwarf stars that can be used as a new set of benchmarks. All the fundamental stellar parameters were based on consistently combining interferometric observations, 3D limb-darkening modelling, and spectroscopic analysis. The next paper in this series will extend our sample to giants in the metal-rich range. 

   more » « less
4. Vulnerability Detection with Code Language Models: How Far Are We?

   Ding, Yangruibo; Fu, Yanjun; Ibrahim, Omniyyah; Sitawarin, Chawin; Chen, Xinyun; Alomair, Basel; Wagner, David; Ray, Baishakhi; Chen, Yizheng (April 2025, 47th International Conference on Software Engineering)

   In the context of the rising interest in code language models (code LMs) and vulnerability detection, we study the effectiveness of code LMs for detecting vulnerabilities. Our analysis reveals significant shortcomings in existing vulnerability datasets, including poor data quality, low label accuracy, and high duplication rates, leading to unreliable model performance in realistic vulnerability detection scenarios. Additionally, the evaluation methods used with these datasets are not representative of real-world vulnerability detection. To address these challenges, we introduce PRIMEVUL, a new dataset for training and evaluating code LMs for vulnerability detection. PRIMEVUL incorporates a novel set of data labeling techniques that achieve comparable label accuracy to humanverified benchmarks while significantly expanding the dataset. It also implements a rigorous data de-duplication and chronological data splitting strategy to mitigate data leakage issues, alongside introducing more realistic evaluation metrics and settings. This comprehensive approach aims to provide a more accurate assessment of code LMs’ performance in real-world conditions. Evaluating code LMs on PRIMEVUL reveals that existing benchmarks significantly overestimate the performance of these models. For instance, a state-of-the-art 7B model scored 68.26% F1 on BigVul but only 3.09% F1 on PRIMEVUL. Attempts to improve performance through advanced training techniques and larger models like GPT-3.5 and GPT-4 were unsuccessful, with results akin to random guessing in the most stringent settings. These findings underscore the considerable gap between current capabilities and the practical requirements for deploying code LMs in security roles, highlighting the need for more innovative research in this domain. 

   more » « less
5. Simulation of Stark-broadened Hydrogen Balmer-line Shapes for DA White Dwarf Synthetic Spectra

   https://doi.org/10.3847/1538-4357/ac4df3  

   Cho, P. B.; Gomez, T. A.; Montgomery, M. H.; Dunlap, B. H.; Fitz Axen, M.; Hobbs, B.; Hubeny, I.; Winget, D. E. (March 2022, The Astrophysical Journal)

   Abstract White dwarfs (WDs) are useful across a wide range of astrophysical contexts. The appropriate interpretation of their spectra relies on the accuracy of WD atmosphere models. One essential ingredient of atmosphere models is the theory used for the broadening of spectral lines. To date, the models have relied on Vidal et al., known as the unified theory of line broadening (VCS). There have since been advancements in the theory; however, the calculations used in model atmosphere codes have only received minor updates. Meanwhile, advances in instrumentation and data have uncovered indications of inaccuracies: spectroscopic temperatures are roughly 10% higher and spectroscopic masses are roughly 0.1M_⊙higher than their photometric counterparts. The evidence suggests that VCS-based treatments of line profiles may be at least partly responsible. Gomez et al. developed a simulation-based line-profile code Xenomorph using an improved theoretical treatment that can be used to inform questions around the discrepancy. However, the code required revisions to sufficiently decrease noise for use in model spectra and to make it computationally tractable and physically realistic. In particular, we investigate three additional physical effects that are not captured in the VCS calculations: ion dynamics, higher-order multipole expansion, and an expanded basis set. We also implement a simulation-based approach to occupation probability. The present study limits the scope to the first three hydrogen Balmer transitions (Hα, Hβ, and Hγ). We find that screening effects and occupation probability have the largest effects on the line shapes and will likely have important consequences in stellar synthetic spectra. 

   more » « less