Searcharxiv⌕ Search

arXiv subjects

Yang Huang

Publications and source records attributed to Yang Huang.

At least 19 recordsLinked to original sources

Non-monotonic Age-metallicity Relations Across the Galactic Disk Revealed by APOGEE Red Clump Stars

Using APOGEE red clump stars and open clusters with precise ages, we investigate the age--metallicity relation (AMR) across the Galactic disk as a function of Galactocentric radius $R$ and vertical height $|Z|$. For stars older than $\sim$6--8 Gyr, the AMR shows little dependence on either $R$ or $|Z|$ within the uncertainties, suggesting rapid and well-mixed chemical evolution in the early disk. In contrast, younger populations exhibit a clear negative radial metallicity gradient, with the inner disk being more metal-rich than the outer disk, consistent with inside-out disk growth since $\sim$6--8 Gyr ago. Most importantly, at fixed location, the AMR for ages $\lesssim6$--8 Gyr is non-monotonic: [Fe/H] decreases toward younger ages, reaches a minimum at intermediate ages, and then rises again. Beyond $R\sim11$\,kpc, the relation becomes even more complex, exhibiting an additional reversal and a distinct metallicity peak at an age of $\sim2$\,Gyr. Such complex behavior cannot be explained by continuous local enrichment alone, but instead points to the combined effects of a multi-phase gas-accretion history and radial migration. Episodes of metal-poor gas inflow can dilute the interstellar medium and reset its chemical evolution, while subsequent enrichment raises the metallicity again; radial migration further reshapes the observed AMR by mixing stellar populations formed at different radii and metallicities. A toy model incorporating inside-out growth, two gas-inflow episodes, and radial migration qualitatively reproduces the observed trends, highlighting the combined roles of episodic gas accretion and radial migration in shaping the chemical evolution of the Galactic disk.

astro-ph.GA↗

Discovery of a likely supermassive neutron star in a remarkably wide binary

Supermassive neutron stars ($M \gtrsim 2\ M_{\odot}$) provide stringent constraints on the equation of state of dense matter. We report the discovery of a wide binary system composed of a red giant star and an invisible companion with an orbital period $P\approx 1218\pm3$ days and a comparatively low eccentricity $e\approx 0.09\pm0.01$ (with $1 σ$ uncertainty). By combining radial velocity measurements from the Large Sky Area Multi-Object Fiber Spectroscopic Telescope and the Apache Point Observatory Galactic Evolution Experiment (APOGEE) with astrometric data from {\it Gaia} DR2 and DR3 catalogues, we derive a Keplerian orbit with an inclination angle $i\approx73^{\circ}\pm11^{\circ}$. The unseen companion has a mass in the range of 1.8--2.8 $M_{\odot}$ (spanning the full spread of all estimates), with a best-fit value of $2.20^{+0.14}_{-0.13}$ $M_{\odot}$. This potentially places the unseen companion within the mass gap between neutron stars and black holes. Spectral disentangling using high-resolution near-infrared APOGEE spectra shows no absorption features from an additional component, suggesting the presence of a compact object: either a supermassive neutron star or a low-mass black hole. The wide, low-eccentricity orbit is difficult to explain through standard binary evolution channels, pointing to the need for improved theoretical models to account for such systems.

astro-ph.HE↗

LAMOST III Low-resolution Spectroscopic Survey: Overview and Target Selection

We present an overview of the LAMOST III Low-resolution Spectroscopic Survey (LRS), including its scientific goals, target selection strategy, and current observing status. Targets in the main survey are selected via weighted random sampling, using the inverse local density in the Hertzsprung-Russell (HR) diagram and in the G-magnitude distribution, with additional enhancement in selected regions of the HR diagram. The main survey is complemented by 35 dedicated add-on catalogs (including very metal-poor star candidates, halo stars, hot/young stars, wide binaries, extragalactic sources and so on) and a time-domain program providing multi-epoch spectroscopy of binaries and other peculiar systems. LAMOST III LRS is designed to map the structure and evolution of the Galactic disk, reconstruct the assembly and early chemical-enrichment history of the Galactic halo, improve stellar-parameter calibration, and characterize rare and variable objects. By June 2025, the LAMOST III LRS had obtained 1.53 million low-resolution spectra from 658 plates, corresponding to 1,264,052 unique sources. Relative to LAMOST I+II, the current sample is enriched in O/B stars, subgiants, main-sequence turn-off stars, metal-poor stars, and other previously underrepresented populations. In particular, the fractions of metal-poor and very metal-poor stars increase from 3.8% and 0.4% in LAMOST I+II to 9.0% and 1.1% in LAMOST III LRS, respectively, demonstrating the improved efficiency of the revised target-selection strategy for Galactic archaeology and stellar astrophysics.

astro-ph.GA↗

A Soft Sponge-Pillar Electrode Platform for Motion-Tolerant Ambulatory EEG and Sleep Monitoring

Long-term wearable electrophysiological recording systems are essential for both clinical and real-world environments. While a wide range of wearable devices have been reported for electrocardiogram (ECG) and electromyogram (EMG) recording applications, most electroencephalography (EEG) measurements still require subjects to remain stationary while wearing rigid and uncomfortable equipment. Here, a soft and wearable sponge pillar EEG electrode is presented for long-term and motion-tolerant EEG recording. The electrode is composed of two layers of a carbon black (CB)-polydimethylsiloxane (PDMS) composite, where a firmer solid base assists in bypassing hair, enabling stable access to the scalp, while a porous sponge tip ensures comfort, reliable contact, and low skin-electrode impedance. Fabricated via a sugar templating process without microfabrication steps, the microporous sponge structure offers low modulus and significantly enhances the contact area with the skin when loaded with conductive gel. This allows the electrode to achieve a lower contact impedance of $1.62 \times 10^{4}\,Ω$ at 10 Hz, outperforming conventional gelled electrodes while maintaining gel stability for over 18 hours. With the electrodes integrated into wearable form factors such as headbands, headphones, and baseball caps, EEG signal recording during motion have been demonstrated, achieving a signal-to-noise ratio (SNR) of 16.46 dB. These features establish the sponge pillar electrode as a promising platform for next generation wearable EEG systems, enabling comfortable, stable, and prolonged brain signal acquisition on ambulatory subjects and in naturalistic environments.

physics.med-ph↗

Measuring Milky Way Mass with RR Lyraes and Blue Horizontal Branch Stars

Measuring the mass of the Milky Way is essential for understanding galaxy formation and evolution, characterizing the Galactic potential and assembly history, and probing the nature of dark matter. We derive the enclosed mass profile of the Milky Way out to Galactocentric distances of $\sim80\,\mathrm{kpc}$ using two halo tracers with full 3D kinematics: RR Lyraes (RRLs) and blue horizontal-branch stars (BHBs). Our sample consists of 7,550 RRLs and 8,230 BHBs constructed from the Large Sky Area Multi-Object Fiber Spectroscopic Telescope (LAMOST), the Sloan Digital Sky Survey (SDSS), and the Dark Energy Spectroscopic Instrument (DESI), spanning Galactocentric distances from $10$ to $100\,\mathrm{kpc}$. We estimate the enclosed mass using two independent methods: the spherical Jeans equation and the tracer mass estimator (TME). The two tracers yield consistent results, with $M(<81.7\,\mathrm{kpc}) = (4.76 \pm 0.23)\times10^{11}\,\mathrm{M}_\odot$ from RRLs and $M(<82.8\,\mathrm{kpc}) = (4.66 \pm 0.35)\times10^{11}\,\mathrm{M}_\odot$ from BHBs. Adopting an NFW profile to extrapolate the enclosed mass beyond the outermost radius probed by our tracers, we infer virial masses of $M_{200} = 8.12^{+0.72}_{-0.65}\times10^{11}\,\mathrm{M}_\odot$ from RRLs and $M_{200} = 8.13^{+0.78}_{-0.67}\times10^{11}\,\mathrm{M}_\odot$ from BHBs. We further validate our methodology using mock RRL and BHB samples constructed from two Auriga halos, finding that the inferred masses are underestimated by $15.5\%\pm2.8\%$ for RRLs and $11.9\%\pm3.4\%$ for BHBs, respectively. After correcting for these systematic errors, the Milky Way virial masses are $M_{200} = 9.79^{+0.93}_{-0.84}\times10^{11}\,\mathrm{M}_\odot$ from RRLs and $M_{200} = 9.24^{+0.95}_{-0.84}\times10^{11}\,\mathrm{M}_\odot$ from BHBs.

astro-ph.GA↗

Observational evidence of the Gaia-Enceladus-Sausage merger's influence on the opposite side of the Galactic Center

The Gaia-Enceladus-Sausage (GES) merger, which occurred approximately 8-11 Gyr ago, is a key event that shaped the Galactic inner halo. Its remnants have been studied extensively and are known to dominate the halo on the Near-side of the Galactic Center (GC), i.e. the side facing the Sun, but systematic evidence for whether the Far-side (the side beyond the GC as seen from the Sun) is similarly affected has been lacking. Using 14,056 red giant stars selected from SMSS DR4 and Gaia DR3 through stringent quality cuts, we compare the metallicity distributions, velocity-space distributions, velocity dispersions, anisotropy parameters, and action-space properties of the Near- and Far-sides. We find clear observational evidence for the influence of GES on the Far-side: it hosts the same metal-rich component peaked at [Fe/H] ~ -1.4, and its metal-rich population (-1.7 < [Fe/H] <= -1.3) shows GES-typical features consistent with the Near-side, namely a "sausage"-shaped velocity distribution and a velocity anisotropy as high as beta ~ 0.8-0.9. Action-space analysis shows that 10.86% +/- 0.23% of Far-side stars carry the GES dynamical imprint, consistent with 11.56% +/- 0.16% on the Near-side. The Far-side shows a residual prograde rotation that persists under increasingly aggressive disk-removal cuts, indicating a genuine kinematic signal rather than disk contamination, and a battery of robustness tests confirms that the symmetry is not an artifact of measurement uncertainties, residual disk contamination, or incomplete Sagittarius-stream removal. These results demonstrate that GES shaped the entire inner halo (r < 28 kpc) globally and symmetrically, with its imprint clearly present on the Far-side as well as the Near-side.

astro-ph.GA↗

Star formation history of Draco revealed by DESI DR1: suppressed early star formation efficiency for Milky Way dwarf galaxies

Dwarf galaxies serve as fossils for understanding the early star formation history of the local Universe. In this work, we use stellar abundances for individual stars derived from the first data release of the Dark Energy Spectroscopic Instrument (DESI DR1) to investigate the star formation history of the dwarf spheroidal galaxy Draco. We simultaneously fit the stellar metallicity distribution function (MDF) and the [Mg/Fe]-[Fe/H] trend using the galactic chemical evolution (GCE) model flexCE. Our best-fit models imply a low star formation efficiency (SFE) of $0.003$ to $0.007\,\mathrm{Gyr}^{-1}$, which is about ten times lower than values reported in the literature, but consistent with recent SFE estimates for ultra-faint dwarf galaxies. By combining recent literature measurements, we find that the SFEs of Milky Way dwarf galaxies scale with stellar mass as $\mathrm{SFE}\propto M_\star^{0.24}$. A low star formation efficiency is associated with a strong outflow efficiency; we estimate a large wind mass-loading factor $η_{\mathrm{wind}}$ of approximately 172-260. This is well in line with the observed outflow-stellar mass relation and with the Feedback in Realistic Environment (FIRE) cosmological simulations. Our results demonstrate the power of large-scale, low-to-medium resolution stellar spectroscopic surveys to deliver robust constraints on the early formation history of galaxies in the local Universe.

astro-ph.GA↗

Scattering of scalar waves by scalarized black holes in Einstein-scalar-Gauss-Bonnet gravity

We systematically investigate the scattering and absorption of a massive scalar field by scalarized black holes in Einstein-scalar-Gauss-Bonnet gravity with quartic coupling $f(ϕ)=α(ϕ^{2}+βϕ^{4})$ and exponential coupling $f(ϕ)=α(1-\operatorname{e}^{-βϕ^{2}})$. The background solutions are obtained numerically via a Chebyshev pseudospectral method, enabling a partial-wave analysis on purely numerical black hole backgrounds. The partial-wave results are cross-validated against geodesic and glory approximations and the Schwarzschild limit with excellent agreements. We find that scalarization suppresses the total absorption cross section relative to the Schwarzschild case, and produces distinctive signatures in the differential scattering cross section. A qualitative difference between the two coupling models is identified: for the quartic coupling model, deviations from Schwarzschild are concentrated in the intermediate-angle region $\sim(70^{\circ},140^{\circ})$, with comparatively small changes at forward angles and near the glory peak; for the exponential coupling model, deviations appear across the full angular range and substantially modify the glory peak, which in some parameter regimes can even exceed the Schwarzschild value. This distinctive angular pattern provides a potential diagnostic for distinguishing EsGB coupling functions from scattering observables.

gr-qc↗

The Nineteenth Data Release of the Sloan Digital Sky Survey

Mapping the local and distant Universe is key to our understanding of it. For decades, the Sloan Digital Sky Survey (SDSS) has made a concerted effort to map millions of celestial objects to constrain the physical processes that govern our Universe. The most recent and fifth generation of SDSS (SDSS-V) is organized into three scientific ``mappers". Milky Way Mapper (MWM) that aims to chart the various components of the Milky Way and constrain its formation and assembly, Black Hole Mapper (BHM), which focuses on understanding supermassive black holes in distant galaxies across the Universe, and Local Volume Mapper (LVM), which uses integral field spectroscopy to map the ionized interstellar medium in the local group. This paper describes and outlines the scope and content for the nineteenth data release (DR19) of SDSS and the most substantial to date in SDSS-V. DR19 is the first to contain data from all three mappers. Additionally, we also describe nine value added catalogs (VACs) that enhance the science that can be conducted with the SDSS-V data. Finally, we discuss how to access SDSS DR19 and provide illustrative examples and tutorials.

astro-ph.GA↗

NucleoNet: A Nuclear-Aware Determinant Ansatz for Multi-Geometry Neural Variational Monte Carlo

Neural-network variational Monte Carlo provides an accurate route to solving the electronic Schrödinger equation, but many neural wavefunctions are optimized for a single nuclear geometry, requiring separate calculations to map out a potential energy surface. We introduce NucleoNet, a nuclear-aware determinant ansatz in which electrons and nuclei form a unified particle set propagated through a shared electron--nucleus backbone that jointly evolves one-particle and two-particle representations. Nuclei carry learned representations that evolve alongside the electronic representations and condition orbital construction, while antisymmetry is imposed only by electronic determinants. On equilibrium small-molecule benchmarks, NucleoNet achieves accuracy comparable to established neural-wavefunction baselines with substantially fewer parameters. For each molecular system, a single model optimized jointly across geometries describes the N$_2$ dissociation curve, Li displacement in LiCN, and the ethane and formamide minimum-energy paths for bond dissociation. Sparse-geometry training generalizes to unseen geometries for N$_2$ and LiCN without further optimization, and N$_2$ ablations show that nuclear streams substantially improve this interpolation. These results show that a compact wavefunction with jointly evolved nuclear representations can describe the geometries of a molecular system with a single set of parameters.

physics.chem-ph↗

Mass Distribution of M31 from its Rotation Curve up to 167 kpc

The mass distribution of the Andromeda galaxy (M31) is fundamental for understanding its dark matter content, assembly history, and dynamical role within the Local Group. However, estimates of M31's total mass remain quite uncertain, partly because of limited dynamical constraints at large galactocentric radii. Moreover, M31's recent merger history has produced abundant tidal substructures and may have left its outer halo out of dynamical equilibrium, complicating equilibrium-based mass estimates. We use over 8,500 tracers with line-of-sight velocities to constrain the rotation curve (RC) of M31 over R=2--167 kpc. In the disk, we derive the RC from a clean tracer sample using the Jeans equation and find good agreement with previous H I measurements. In the halo, we use Gaussian mixture modeling to account for tidal substructures and Galactic foreground contamination and constrain the kinematics of the residual dynamically hot halo. Assuming this halo population can be treated as an equilibrium tracer, we infer the outer RC and constrain M31's mass distribution for different velocity anisotropies beta and dark matter halo profiles. The NFW and Dekel--Zhao (DZ) models yield broadly consistent total masses ranging from 1.12(+0.11/-0.10) to 1.63(+0.21/-0.17)x10^12 Msun for beta=0.3--0.7. In contrast, the Einasto profile favors substantially lower masses of 0.41(+0.09/-0.06) to 0.86(+0.21/-0.16)x10^12 Msun, highlighting significant dependence on the adopted halo profile, although Einasto solutions are more in tension with independent mass estimates based on M31 satellite galaxies than the NFW and DZ models at fixed beta. Overall, our analysis provides a comprehensive measurement of M31's mass distribution under the assumption of dynamical equilibrium, while also quantifying the systematic uncertainties associated with halo velocity anisotropy and the choice of dark matter density profile.

astro-ph.GA↗

Distorting Kerr Images with Parity-Odd Scalar Hair

We investigate thin-disk imaging of Kerr black holes with synchronized scalar hair, focusing on backreacted parity-odd excited states of a complex scalar field minimally coupled to Einstein gravity. The spacetime displays a core-double-torus lensing structure, with a central black hole surrounded by two scalar clouds. We study the dependence of the images on hair strength and viewing angle, identifying a weak-hair regime close to Kerr. With increasing hair, the photon ring and shadow region shrink and become more distorted. In the strong-hair regime, gravitational lensing produces new features, including multiple disconnected shadow components, crescent-shaped structures, and signatures of chaotic lensing. For nearly edge-on viewing angles, repeated equatorial crossings generate nested ring-like patterns. These results highlight possible geometric signatures of black holes with excited scalar hair.

gr-qc↗

High Resolution Spectroscopic Follow-up Observation Results for 13 EMP Candidates Selected by Narrow-band Photometry

Extremely metal-poor (EMP; [Fe/H] < -3.0) stars preserve key information about the earliest stages of Galactic chemical evolution. Based on a training set of spectroscopic metallicities, we constructed catalogs of more than 120,000 EMP candidates from large-scale narrow-band photometric surveys. To validate this approach, we obtained high-resolution CFHT/ESPaDOnS spectra for 13 candidates selected from the SkyMapper-based catalog. The follow-up observation confirms that the narrow-band photometric selection is effective in identifying very metal-poor stars and retaining a substantial EMP fraction: four targets are confirmed as EMP stars, and all targets remain very metal-poor with [Fe/H] < -2.4. The photometric metallicities are systematically lower than the spectroscopic values by 0.44 dex, with a relatively small scatter of 0.19 dex, indicating a systematic offset in the photometric metallicity scale that could be reduced through improved calibration. Abundances for more than 20 chemical species are derived, leading to the identification of a new potassium-enhanced r-II star, J2001-1215. By combining chemical abundances with orbital properties, five stars are found to be dynamically consistent with known Galactic substructures, including Gaia--Sausage--Enceladus, Thamnos, and Sequoia. These stars provide high-resolution chemical measurements for the extremely metal-poor regime of these Galactic substructures. These results underscore the role of narrow-band photometry in efficiently selecting EMP candidates for targeted high-resolution spectroscopic follow-up, thereby enabling detailed chemical and dynamical studies of the early Milky Way.

astro-ph.GA↗

Decoding RR Lyrae light curves with deep learning for accurate absolute magnitude estimation

RR Lyrae stars are essential standard candles for distance measurements in the Milky Way and nearby galaxies. Traditional estimates rely on the Period--Absolute Magnitude--Metallicity relation but are limited by uncertainties in metallicity determinations. We present a deep learning approach that directly predicts absolute magnitudes from RRab and RRc light curves, eliminating the need for metallicity estimates. Our model achieves validation precisions of 0.053 mag and 0.036 mag (approximately 2.5% and 1.7% in distance) for individual RRab and RRc stars, respectively. Tests on globular clusters yield typical distance precisions of 1.0% for RRab and 1.7% for RRc stars. For the benchmark systems, combining the RRab- and RRc-based models yields distance moduli of 18.498 $\pm$ 0.001$_{\text{stat}}$ $\pm$ 0.018$_{\text{sys}}$ mag for the Large Magellanic Cloud and 19.564 $\pm$ 0.003$_{\text{stat}}$ $\pm$ 0.019$_{\text{sys}}$ mag for the Sculptor dwarf spheroidal galaxy. These measurements are in excellent agreement with previous results, achieve a distance precision of approximately 1%, and represent a 1.8-fold improvement over traditional RR Lyrae calibration relations. Our approach showcases the ability of AI to directly extract key physical parameters from complex, information-rich light curves, resolve degeneracies, and scale to broader applications.

astro-ph.GA↗

Radio Emission with Dust: A Bow-Shock Interpretation for the TDE Candidate AT 2019avd

Forward shocks produced by interactions between a single ejected blob and the circumnuclear medium (CNM) around a supermassive black hole are widely invoked to explain radio emission from tidal disruption events (TDEs). However, recent observational evidence such as rapidly rising radio emission, double-peaked broadband spectral energy distributions, and declining shock energies at late times poses challenges to this picture. Alternative scenarios have therefore been proposed, including bow shocks arising from outflows colliding with dense clouds, the production of new ejecta at late times, and variations in microphysical parameters. To investigate the roles of these factors, we analyze the long-lived radio flare of the TDE candidate AT 2019avd, for which dust has been identified through mid-infrared observations, using a radio dataset spanning more than five years. We find that a bow-shock scenario can account for the majority of the observed radio emission, while a forward shock, or an additional bow shock involving different clouds, may begin to dominate at late times. However, a forward-shock scenario remains viable when allowing for evolving microphysical parameters and more complex CNM structures. We discuss the remaining tensions in both the outflow-CNM and outflow-cloud interaction scenarios.

astro-ph.HE↗

Fourier Neural Operators for Composition-Driven Crystal Structure Discovery

Crystalline materials discovery is essential for energy, electronics, and catalysis, but the vast chemical and structural space makes exhaustive screening infeasible. Existing voxel-based methods are limited by the local receptive fields of three-dimensional convolutional neural networks and the posterior collapse of high-dimensional variational autoencoders. Here, we develop a Fourier Neural Operator (FNO)-based crystal-field solver that maps a prescribed chemical formula and lattice parameters to periodic number-density and electron-density fields. By operating on global Fourier modes, the solver captures long-range correlations in periodic crystal fields beyond conventional local convolutions. Building on this solver, we construct a coupled generation-solving framework in which a conditional variational autoencoder generates diverse candidate lattice parameters in a low-dimensional basis-coefficient space, followed by density-field prediction and atomic reconstruction through peak detection, position optimization, and weight optimization. The reconstructed structures are further screened using voxel-level filtering, machine-learning interatomic-potential relaxation, and first-principle calculations. The framework generates novel structures across 104 chemical formulas with competitive reconstruction accuracy, demonstrating high generative diversity and structural validity. By extending Fourier neural operators to periodic crystal fields and coupling them with composition-conditioned lattice generation, our approach provides a scalable route to crystal structure discovery from prescribed chemical compositions.

cond-mat.mtrl-sci↗

Evolution of Stellar Activity and Habitable Zone (EATEN): III. X-ray Activity of Dwarfs in Open Clusters and Field Stars

Stellar X-ray emission serves as a direct diagnostic of coronal activity, which is fundamentally linked to coronal heating processes. It also strongly influences the atmospheres and long-term habitability of orbiting exoplanets. Investigating how this high-energy emission evolves is therefore essential for understanding the evolution of stellar magnetic dynamos and planetary atmospheres and habitability. In this work, we investigate the evolution of X-ray activity and XUV irradiation for a sample of F-M dwarf stars based on Chandra and XMM-Newton observations. We find that F- and G-type stars broadly follow the traditional evolutionary picture of an early saturated (or weakly declining) phase followed by a modest decline, whereas K- and M-type stars exhibit a clear three-phase evolution of a saturated phase, an intermediate phase of rapid decay, and a final modest decline phase. By combining X-ray, ultraviolet, and Ca II H&K bands, we show that coronal emission becomes increasingly dominant toward lower-mass stars. Based on the cumulative XUV emission calculated from our fitted relation, planets around F- and G-type stars experience relatively moderate XUV environments, while those around K- and M-type stars may exceed the empirical cosmic shoreline shortly after reaching the main sequence, though this conclusion depends on the adopted shoreline value.

astro-ph.SR↗

Cascading Relevance-driven Recommendation Network for CTR Prediction in Trigger-Introduced Recommendation

E-commerce has emerged as crucial platforms for people's daily consumption and shopping interests. There is a new recommendation scenario, Trigger-Introduced Recommendation (TIR), where users click interested product, which is defined as the trigger item, containing their instant interest, and in the undertaking page following the relevant target items. Distinguished from traditional search and recommendation scenarios, trigger contains relatively strong instant interest, which is more vague and implicit compared to search terms. Relying on large amounts of labeled data, existing methods lack the exploration of trigger relevance, which affects users' immersive experience. To alleviate this problem, we propose the Cascading Relevance-driven Recommendation Network (CRRN) to emphasize the interaction and relevance between trigger and target, comprising three essential components: 1) the Trigger-Target Interaction layer extracts interaction features of trigger and target based on personalized gating. 2) Cascading Interest Fusion module explicitly estimates users' trigger intention and fuses instant and personalized interests adaptively with cascading attention blocks. 3) Category-assisted Pairwise Loss enhances trigger relevance with the guidance of category association between trigger and target. Extensive experiment results show that CRRN outperforms recent state-of-the-art methods on both industrial and public datasets. Online A/B tests further validate the effectiveness of our method. Our code is available at https://github.com/a-little-cabbage/CRRN.

cs.IR↗