arXiv subjects

Yang Sun

Publications and source records attributed to Yang Sun.

At least 91 records · Page 5Linked to original sources

arXiv2025

Confirming Near- to Mid-IR Photometrically-Identified Obscured AGNs in the JWST era

We evaluate the underlying assumptions for the identification of Active Galactic Nuclei (AGNs) through near- and mid-infrared photometry and spectral energy distribution (SED) fitting out to z ~ 3. For massive galaxies, log(M) > 9.5, our high resolution spectra of the rest optical range generally confirm the results of SED fitting, which relies primarily on excesses above the stellar emission between 1 and 6 microns to identify AGN. However, the method is undermined if the redshift used for the SED fitting is incorrect. Low mass galaxies, log(M) < 9:5, can contain relatively warm dust that emits in the 4 - 6 micron range. We show that the potential contamination of AGN samples by purely star forming low-mass galaxies can be avoided by the use of the infrared properties of Haro 11 as a limiting star-forming SED template. However, relatively few star forming galaxies emit as strongly in the 3 - 6 micron range as this template, so this could result in missing some obscured AGNs to avoid a minor contamination. Including the behavior of the galaxies at rest lamda ~ 13.5 microns can mitigate this problem and yield more complete samples of bona fide AGN. JWST/MIRI supports this approach out to z ~ 0.6.

astro-ph.GA↗

arXiv2025

Graph-Reward-SQL: Execution-Free Reinforcement Learning for Text-to-SQL via Graph Matching and Stepwise Reward

Reinforcement learning (RL) has been widely adopted to enhance the performance of large language models (LLMs) on Text-to-SQL tasks. However, existing methods often rely on execution-based or LLM-based Bradley-Terry reward models. The former suffers from high execution latency caused by repeated database calls, whereas the latter imposes substantial GPU memory overhead, both of which significantly hinder the efficiency and scalability of RL pipelines. To this end, we propose a novel reward model framework for RL-based Text-to-SQL named Graph-Reward-SQL, which employs the GMNScore outcome reward model. We leverage SQL graph representations to provide accurate reward signals while significantly reducing time cost and GPU memory usage. Building on this foundation, we further introduce StepRTM, a stepwise reward model that provides intermediate supervision over Common Table Expression (CTE) subqueries. This encourages both functional correctness and readability of SQL. Extensive comparative and ablation experiments on standard benchmarks, including Spider and BIRD, demonstrate that our method consistently outperforms existing reward models.

cs.LG↗

arXiv2025

JADES: The Star Formation and Dust Attenuation Properties of Galaxies at 3<z<7

We present the star formation and dust attenuation properties for a sample of 602 galaxies at redshifts $\rm{3<z<7}$, as part of the JADES survey. Our analysis is based on measurements of the $\rm{H}α/\rm{H}β$ Balmer Decrement using medium resolution (R$\sim$1000) spectroscopic observations with the JWST/NIRSpec Micro-Shutter Assembly. Stellar masses and star formation rates (SFRs) are inferred with \texttt{Prospector} using deep multi-band imaging. We utilize the Balmer decrement to measure dust-corrected H$α$-based SFRs, taking into account the subsolar metallicities observed in galaxies at high redshift. We confirm, with our large sample size, that the correlation between the Balmer decrement and stellar mass is already established out to $z\sim7$. We find that the relation between the Balmer decrement and stellar mass does not significantly evolve from the local universe to $z\sim7$. We investigate the UV slope as a function of the Balmer optical depth and find that the best-fit correlation for our high redshift sample is sSFR dependent and significantly different at high redshift when compared to galaxies at $z\approx 0$ and $z \approx 2$. For the highest sSFR galaxies in our sample, there is no significant correlation between the UV slope and Balmer optical depth. This is evidence that the UV slope should be used with great caution to correct for dust in high redshift galaxies.

astro-ph.GA↗

arXiv2025

Effective $β$-decay rates of $r$-process waiting points in realistic stellar environments

Reliable nuclear weak rates are key inputs for understanding the origin of heavy elements and constraining the environments of the corresponding stellar nucleosynthesis. We present the effective stellar $β^-$-decay rates of the $N=50, 82, 126$ $r$-process waiting-point nuclei in realistic stellar environments with high temperature, high density and strong magnetic field. Both allowed and first-forbidden transitions are considered, and transitions from the low-lying states of parent nuclei due to the thermal population are taken into account properly. The stellar $β^-$-decay rates of the $N=50, 82$ waiting points are not sensitive to stellar temperature, while those of the $N=126$ waiting points increase rapidly with stellar temperature. With the increase of stellar density, the electron chemical potential increases accordingly, which leads to reduction of the stellar $β$-decay rates. Besides, the stellar $β$-decay rates are found to increase rapidly with the magnetic field $B$ when $B \gtrsim 10^{14}$ G. Depending on the stellar temperature, density and magnetic field, the rates may vary by several orders of magnitude, which indicates that dynamic $β$-decay rates for corresponding stellar conditions may be indispensable inputs for understanding the $r$-process nucleosynthesis.

nucl-th↗

arXiv2025

Prediction of Li3Fe8B8 compound with rapid one-dimensional ion diffusion channels

Using a computational crystal structure search in the Li-Fe-B ternary system, we predict a stable phase of Li3Fe8B8, featuring 1D channels that enable rapid Li-ion transport. Ab initio molecular dynamics simulations show that the Li-ion diffusion coefficient in Li3Fe8B8 surpasses that of common electrode and conductive additive materials by several orders of magnitude. The high diffusion in Li3Fe8B8 can be explained by the Frenkel-Kontorova model, which describes an incommensurate state between the Li diffusion chain and the periodic potential field caused by the FeB backbone structure. The favorable lithium-ion diffusivity and mechanical properties of Li3Fe8B8 make it a promising conductive additive for battery materials. Its itinerant ferromagnetism also offers a platform for exploring correlated-electron magnetism and spin-dependent phenomena.

cond-mat.mtrl-sci↗

arXiv2025

The Fe-Ni phase diagram and the Earth's inner core structure

The Fe-Ni alloy is believed to be the main component of Earth's core. Yet, a comprehensive understanding of phase equilibria near the melting point of this alloy under core conditions is still lacking, leaving the effect of nickel inconclusive. Using ab initio simulations, we computed Gibbs free energy and phase diagram for liquid and solid solutions of the Fe-Ni alloy under conditions close to the inner core, considering inner-shell electron contributions and non-ideal mixing effects. The Fe-Ni phase diagram provides crucial insights for understanding previous experimental observations and crystallization simulations of the Fe-Ni alloy under core conditions. It also presents new scenarios for inner core structures, suggesting bcc-liquid coexistence at the inner core boundary and the possibility of multi-layer structures consisting of bcc-hcp composites within the inner core. Our work clarifies nickel's substantial impact on the inner core structure, providing new constraints for the study of core's composition and formation.

physics.geo-ph↗

arXiv2025

The $z=7.08$ quasar ULAS J1120+0641 May Never Reach a "Normal" Black Hole to Stellar Mass Ratio

JWST observations of quasars in the Epoch of Reionization have revealed that many lie in host galaxies that are severely undermassive relative to the supermassive black holes. It is unclear how these systems will evolve to the tight local relation between stellar mass and black hole mass. We search for companions around the z=7.08 quasar ULAS J1120+0641 using JWST/NIRCam narrow, medium, and wide-band photometry to identify [O III] emitters at the quasar redshift, and explore the potential for growth of the host galaxy through future mergers. We find 22 sources near the quasar's redshift across our two 4.4 arcmin$^2$ fields, indicating that environment of ULAS J1120+0641 is strongly overdense in z~7.1 galaxies relative to the field. We estimate the potential future mass budget of the quasar host galaxy by summing the current stellar and gas masses of the quasar host and surrounding galaxies, correcting for incompleteness and selection effects. With no further black hole growth, ULAS J1120+0641 is unlikely to reach a $M_{\mathrm{BH}}/M_*$ ratio less than ~2.5% at z=0, still much higher than typical for local galaxies. However, such systems -- a quiescent black hole in a low-luminosity galaxy -- may have escaped detection locally if they are sufficiently distant.

astro-ph.GA↗

arXiv2025

Influence and Connectivity in Networks: A Generating Function Approach

Many widely used network centralities are based on counting walks that meet specific criteria. This paper introduces a systematic framework for walk enumeration using generating functions. We introduce a first-passage decomposition that uniquely divides any walk passing through specified nodes or links into two components: a first-reaching walk and a subsequent walk. This decomposition yields a system of interconnected equations that relate three disjoint categories of walks: unrestricted walks, walks that avoid specific elements, and walks that pass through designated sets. The framework offers a range of applications, including evaluating the effects of structural interventions, such as node or link modifications, on network walks, generalizing target centrality to multi-receiver scenarios in information networks, and comparing different strategies for adding links.

econ.TH↗

arXiv2025

Discovery and Synthesis of a Family of Boride Altermagnets

Borides are a rich material family. To push the boundaries of borides' properties and applications into broader fields, we have conducted systematic theoretical and experimental searches for synthesizable phases in ternary borides TM$_2$B$_2$ (T = 3d, M = 4d/5d transition metals). We find that TM$_2$B$_2$ in the FeMo$_2$B$_2$-type and CoW$_2$B$_2$-type structures form a large family of stable/metastable materials of 120 members. Among them, we identify 40 materials with stable magnetic solutions. Further, we discover 11 altermagnets in the FeMo$_2$B$_2$-type structure. So far, boride altermagnets are rare. In these altermagnets, T = Fe or Mn atoms are arranged in parallel T-chains with strong ferromagnetic intrachain couplings and antiferromagnetic interchain couplings. They simultaneously exhibit electronic band spin splitting, typical of ferromagnetism, and zero net magnetization, typical of antiferromagnetism. They also exhibit magnonic band chiral splitting. Both effects originate from the unique altermagnetic symmetries crucially constrained by the nonmagnetic atoms in the structure. Transport properties of relevance to spintronic applications, including the strain-induced spin-splitter effect and anomalous Hall effect, are predicted. An iodine-assisted synthesis method for TM$_2$B$_2$ is developed, using which 7 of the predicted low-energy phases are experimentally synthesized and characterized, including 4 altermagnets. This work expands the realm of borides by offering new opportunities for studying altermagnetism and altermagnons in borides. It also provides valuable insights into the discovery and design of altermagnets. By demonstrating that altermagnets can exist as families sharing a common motif, this work paves a feasible route for discovering altermagnets by elemental substitutions and high-throughput computations.

cond-mat.mtrl-sci↗

arXiv2025

Structure Maintained Representation Learning Neural Network for Causal Inference

Recent developments in causal inference have greatly shifted the interest from estimating the average treatment effect to the individual treatment effect. In this article, we improve the predictive accuracy of representation learning and adversarial networks in estimating individual treatment effects by introducing a structure keeper which maintains the correlation between the baseline covariates and their corresponding representations in the high dimensional space. We train a discriminator at the end of representation layers to trade off representation balance and information loss. We show that the proposed discriminator minimizes an upper bound of the treatment estimation error. We can address the tradeoff between distribution balance and information loss by considering the correlations between the learned representation space and the original covariate feature space. We conduct extensive experiments with simulated and real-world observational data to show that our proposed Structure Maintained Representation Learning (SMRL) algorithm outperforms state-of-the-art methods. We also demonstrate the algorithms on real electronic health record data from the MIMIC-III database.

stat.ML↗

arXiv2025

Effect of Doping on the phase stability and Superconductivity in LaH10

We present a computational investigation into the effects of chemical doping with 15 different elements on phase stability and superconductivity in the LaH10 structure. Most doping elements were found to induce softening of phonon modes, enhancing electron-phonon coupling and improving critical superconducting temperature while weakening dynamical stability. Unlike these dopants, Ce was found to extend the range of dynamical stability for LaH10 by eliminating the van Hove singularity near the Fermi level. The doped compound, La0.75Ce0.25H10, maintains high-temperature superconductivity. We also demonstrate that different Ce doping configurations in the LaH10 structure have a minimal effect on energetic stability and electron-phonon coupling strength. Our findings suggest that Ce is a promising dopant to stabilize LaH10 at lower pressures while preserving its high-temperature superconductivity.

cond-mat.supr-con↗

arXiv2025

JADES and SAPPHIRES: Galaxy Metamorphosis Amidst a Huge, Luminous Emission-line Region

We report the discovery of a remarkably large and luminous line-emitting nebula extending on either side of the Balmer-break galaxy JADES-GS-518794 at z=5.89, detected with JADES JWST/NIRCam imaging in [O III]$λλ$4959,5007 and H$α$ and spectroscopically confirmed with NIRCam/WFSS thanks to the pure-parallel SAPPHIRES programme. The end-to-end velocity offset is $Δv=830\pm130$ km s$^{-1}$. Nebulae with such large size and high luminosity (25-pkpc diameter, L[O III] = $1.2\times 10^{10}$ L$_\odot$) are routinely observed around bright quasars, unlike JADES-GS-518794. With a stellar mass of $10^{10.1}$ M$_\odot$, this galaxy is at the knee of the mass function at z=6. Its star-formation rate declined for some time (10-100 Myr prior to observation), followed by a recent (10 Myr) upturn. This system is part of a candidate large-scale galaxy overdensity, with an excess of Balmer-break galaxies compared to the field (3 $σ$). We discuss the possible origin of this nebula as material from a merger or gas expelled by an active galactic nucleus (AGN). The symmetry of the nebula, its bubble-like morphology, kinematics, high luminosity, and the extremely high equivalent width of [OIII] together favour the AGN interpretation. Intriguingly, there may be a physical connection between the presence of such a large, luminous nebula and the possible metamorphosis of the central galaxy towards quenching.

astro-ph.GA↗

arXiv2025

DTN: Deep Multiple Task-specific Feature Interactions Network for Multi-Task Recommendation

Neural-based multi-task learning (MTL) has been successfully applied to many recommendation applications. However, these MTL models (e.g., MMoE, PLE) did not consider feature interaction during the optimization, which is crucial for capturing complex high-order features and has been widely used in ranking models for real-world recommender systems. Moreover, through feature importance analysis across various tasks in MTL, we have observed an interesting divergence phenomenon that the same feature can have significantly different importance across different tasks in MTL. To address these issues, we propose Deep Multiple Task-specific Feature Interactions Network (DTN) with a novel model structure design. DTN introduces multiple diversified task-specific feature interaction methods and task-sensitive network in MTL networks, enabling the model to learn task-specific diversified feature interaction representations, which improves the efficiency of joint representation learning in a general setup. We applied DTN to our company's real-world E-commerce recommendation dataset, which consisted of over 6.3 billion samples, the results demonstrated that DTN significantly outperformed state-of-the-art MTL models. Moreover, during online evaluation of DTN in a large-scale E-commerce recommender system, we observed a 3.28% in clicks, a 3.10% increase in orders and a 2.70% increase in GMV (Gross Merchandise Value) compared to the state-of-the-art MTL models. Finally, extensive offline experiments conducted on public benchmark datasets demonstrate that DTN can be applied to various scenarios beyond recommendations, enhancing the performance of ranking models.

cs.IR↗

arXiv2025

CodeContests+: High-Quality Test Case Generation for Competitive Programming

Competitive programming, due to its high reasoning difficulty and precise correctness feedback, has become a key task for both training and evaluating the reasoning capabilities of large language models (LLMs). However, while a large amount of public problem data, such as problem statements and solutions, is available, the test cases of these problems are often difficult to obtain. Therefore, test case generation is a necessary task for building large-scale datasets, and the quality of the test cases directly determines the accuracy of the evaluation. In this paper, we introduce an LLM-based agent system that creates high-quality test cases for competitive programming problems. We apply this system to the CodeContests dataset and propose a new version with improved test cases, named CodeContests+. We evaluated the quality of test cases in CodeContestsPlus. First, we used 1.72 million submissions with pass/fail labels to examine the accuracy of these test cases in evaluation. The results indicated that CodeContests+ achieves significantly higher accuracy than CodeContests, particularly with a notably higher True Positive Rate (TPR). Subsequently, our experiments in LLM Reinforcement Learning (RL) further confirmed that improvements in test case quality yield considerable advantages for RL.

cs.SE↗

arXiv2025

A Systematic Search for Galaxies with Extended Emission Line and Potential Outflows in JADES Medium-Band Images

For the first time, we present a systematic search for galaxies with extended emission line and potential outflow features using \textit{JWST} medium-band images in the GOODS-S field. This is done by comparing the morphology in medium-band images to adjacent continuum and UV bands. We look for galaxies that have a maximum extent 50\% larger, an excess area 30\% greater, or an axis ratio difference of more than 0.3 in the medium band compared to the reference bands. After visual inspection, we find 326 candidate galaxies at $1.4 < z < 8.4$, with a peak in the population near cosmic noon, benefiting from the good coverage of the medium-band filters. By fitting their SEDs, we find that the candidate galaxies are at least 20\% more bursty in their star-forming activity and have 50\% more young stellar populations compared to a control sample selected based on the continuum band flux. Additionally, these candidates exhibit a significantly higher production rate of ionizing photons. We further find that candidates hosting known AGN produce extended emission that is more anisotropic compared to non-AGN candidates. A few of our candidates have been spectroscopically confirmed to have prominent outflow signatures through NIRSpec observations, showcasing the robustness of the photometric selection. Future spectroscopic follow-up will better help verify and characterize the kinematics and chemical properties of these systems.

astro-ph.GA↗

arXiv2025

SMILES: Potentially Higher Ionizing Photon Production Efficiency in Overdense Regions

The topology of reionization and the environments where galaxies efficiently produce ionizing photons are key open questions. For the first time, we investigate the trend between ionizing photon production efficiency, $ξ_{\rm ion}$, and galaxy overdensity, $\log(1+δ)$. We analyze the ionizing properties of 79 galaxies between $1.0 < z < 5.2$ using JWST NIRSpec medium-resolution spectra from the Systematic Mid-infrared Instrument (MIRI) Legacy Extragalactic Survey (SMILES) program. Among these, 67 galaxies have H$α$ coverage, spanning $1.0 < z < 3.1$. The galaxy overdensity, $\log(1+δ)$, is measured using the JADES photometric catalog, which covers the SMILES footprint. For the subset with H$α$ coverage, we find that $\logξ_{\rm ion}$ is positively correlated with $\log(1+δ)$, with a slope of $0.94_{-0.46}^{+0.46}$. Additionally, the mean $ξ_{\rm ion}$ for galaxies in overdense regions ($\log(1+δ) > 0.1$) is 2.43 times that of galaxies in lower density regions ($\log(1+δ) < 0.1$). This strong trend is found to be independent of redshift evolution. Furthermore, our results confirm the robust correlations between $ξ_{\rm ion}$ and the rest-frame equivalent widths of the [O III] or H$α$ emission lines. Our results suggest that galaxies in high-density regions are efficient producers of ionizing photons.

astro-ph.GA↗

arXiv2025

Ferroelasticity, shear modulus softening, and the tetragonal-cubic transition in davemaoite

Davemaoite (Dm), the cubic phase of CaSiO3-perovskite (CaPv), is a major component of the Earth's lower mantle. Understanding its elastic behavior, including its dissolution in bridgmanite (MgSiO3-perovskite), is crucial for interpreting lower mantle seismology. Using machine-learning interatomic potentials and molecular dynamics, we investigate CaPv's elastic properties across the tetragonal-cubic transition. Our equations of state align well with experimental data at 300 K and 2,000 K, demonstrating the predictive accuracy of our trained potential. We simulate the ferroelastic hysteresis loop in tetragonal CaPv, which has yet to be investigated experimentally. We also identify a significant temperature-induced shear modulus softening near the phase transition, characteristic of ferroelastic-paraelastic transitions. Unlike previous elasticity studies, our softening region does not extend to slab geotherm conditions. We suggest that ab initio-quality computations provide a robust benchmark for shear elastic softening associated with ferroelasticity, a challenging property to measure in these materials.

cond-mat.mtrl-sci↗

arXiv2025

DPGP: A Hybrid 2D-3D Dual Path Potential Ghost Probe Zone Prediction Framework for Safe Autonomous Driving

Modern robots must coexist with humans in dense urban environments. A key challenge is the ghost probe problem, where pedestrians or objects unexpectedly rush into traffic paths. This issue affects both autonomous vehicles and human drivers. Existing works propose vehicle-to-everything (V2X) strategies and non-line-of-sight (NLOS) imaging for ghost probe zone detection. However, most require high computational power or specialized hardware, limiting real-world feasibility. Additionally, many methods do not explicitly address this issue. To tackle this, we propose DPGP, a hybrid 2D-3D fusion framework for ghost probe zone prediction using only a monocular camera during training and inference. With unsupervised depth prediction, we observe ghost probe zones align with depth discontinuities, but different depth representations offer varying robustness. To exploit this, we fuse multiple feature embeddings to improve prediction. To validate our approach, we created a 12K-image dataset annotated with ghost probe zones, carefully sourced and cross-checked for accuracy. Experimental results show our framework outperforms existing methods while remaining cost-effective. To our knowledge, this is the first work extending ghost probe zone prediction beyond vehicles, addressing diverse non-vehicle objects. We will open-source our code and dataset for community benefit.

cs.RO↗