arXiv subjects

Lili Yang

Publications and source records attributed to Lili Yang.

At least 19 recordsLinked to original sources

arXiv2026

Reconstruction of Shower-like Events in NEON Using Likelihood and Graph Neural Network Methods

The Neutrino Observatory in the Nanhai (NEON) is a proposed deep-sea neutrino telescope deployed in the South China Sea. Accurate reconstruction of shower-like events is crucial for neutrino energy measurements and multi-messenger astronomy, yet it poses significant challenges due to seawater optical attenuation, irregular detector geometry, and substantial $^{40}\mathrm{K}$ ambient background. In this work, we present the first comprehensive reconstruction framework for shower-like events in NEON, encompassing both a physics-driven maximum likelihood estimation (MLE) method and a data-driven Graph Neural Network (GNN). The traditional MLE framework integrates spatial-isochronic hit selection, vertex reconstruction via time-residual M-estimator minimization, and decoupled directional and energy estimation based on pre-computed photon distribution tables. Physical calibrations, including PMT angular acceptance, hit-level time slewing corrections, and an effective line-source shower extension, are incorporated into the likelihood formulation. In parallel, a two-stage GNN is developed to capture intra-DOM PMT correlations and distance-weighted inter-DOM topological patterns. Simulation studies show that the MLE method achieves an overall median angular resolution of $4.19^\circ$ and an energy resolution of 25\%-37\% over 1 TeV to 1 PeV with negligible systematic bias. The GNN further improves reconstruction fidelity in the low-to-intermediate energy regime, achieving a median angular resolution of $1.8^\circ$ at 30 TeV and an energy resolution of $\sim$ 20\% between 40 and 300 TeV. Based on these reconstruction performances, the effective area and point-source discovery potential of NEON are evaluated. This framework establishes an essential reconstruction benchmark for NEON and provides practical methodologies for future next-generation deep-sea neutrino telescopes.

astro-ph.IM↗

arXiv2026

GenCAR: Generative Counterfactual Alignment with Risk-Controlled Selection for Out-of-Distribution Recommendation

Serving useful recommendations under distribution shift is crucial for balancing utility and risk in out-of-distribution (OOD) recommendation. However, most existing OOD methods improve ranking or construct counterfactual candidates without controlling the proxy-label false discovery rate (FDR) of the served set. In this work, we formulate OOD serving as the $\alpha$-Valid Counterfactual Recommendation ($\alpha$-VCR) problem to retain candidate support learned from counterfactual supervision while controlling proxy-label FDR, and propose GenCAR, which couples preference-grounded counterfactual supervision with calibrated set selection. In particular, GenCAR fixes the stable-preference representation while intervening on the environmental factor, grounds offline large language model proposals through preference anchors and trust-radius filtering, and uses conformal $p$-values for Benjamini--Hochberg selection. We theoretically bound conditional counterfactual approximation error and prove finite-sample, distribution-free control of proxy-label FDR under exchangeability and positive regression dependence, with a Benjamini--Yekutieli guarantee under arbitrary dependence. Extensive experiments audit realized proxy false discovery proportions and demonstrate that GenCAR consistently enhances OOD candidate recovery across diverse benchmarks.

cs.IR↗

arXiv2026

Minkowski and Abraham Momenta Revisited from the Perspective of Photon Dynamics

The distinction between Minkowski and Abraham momenta has traditionally been discussed in the context of electron dynamics. In this work, we revisit the problem from the perspective of photon dynamics in a medium, and show that, within a field-theoretic formulation, their relation is naturally encoded through covariant and ordinary derivatives, together with a medium-induced gauge field. In this quantized framework, the microscopic origin of the optical Hall effect is identified as a spin-Zeeman-type coupling that emerges from the Dirac-like equation for photons.

physics.optics↗

arXiv2026

IRGNN: Efficient Invariant Radar Graph Neural Network for Radar Point Cloud Object Detection

Perception is a fundamental component of autonomous driving systems. While LiDAR-based methods have achieved remarkable progress in object detection, their reliability can degrade under adverse weather conditions. Radar point clouds provide a robust alternative due to their resilience to bad weather and low-illumination scenarios. However, radar point clouds are typically sparse, unordered, and less informative than LiDAR data, making it challenging to directly apply existing LiDAR-based perception methods. To address these challenges, we propose IRGNN, an Invariant Radar Graph Neural Network for radar point cloud object detection. IRGNN first reconstructs radar point clouds into graph representations using translation- and rotation-invariant feature designs, enabling robust modeling of sparse radar measurements. It then employs an improved message passing neural network (MPNN) with residual connections and a virtual node layer to enhance local feature propagation and global context modeling. Finally, task-specific heads are applied to the learned graph representations for object classification and bounding box prediction. Experimental results on the RadarScenes dataset show that IRGNN outperforms existing radar-based object detection methods and achieves competitive performance. In addition, IRGNN significantly reduces computational cost and memory usage during inference, demonstrating its effectiveness and practical potential for efficient radar-based perception in autonomous driving.

cs.CV↗

arXiv2026

ProME: Prototype-Margin Environments with Repair-Aware Selection for Group-Robust Learning

Group-robust learning is crucial for maintaining accuracy on rare subpopulations when training-group labels are unavailable. However, existing methods often infer environments from a separate reference model and select representations before fitting the classifier used at deployment, leaving both decisions misaligned with the deployed predictor. In this work, we formulate group robustness without training-group labels as the endogenous environments with repair-aware selection (ERAS) problem, and propose ProME (Prototype-Margin Environments) to align both decisions with the deployed predictor. ProME splits prototype margins at their median to construct approximately balanced environments along the training trajectory, and fits a group-balanced linear head on group-annotated validation data to rank the resulting predictors by validation worst-group accuracy. We theoretically bound the worst risk across the inferred environments for a fixed predictor and partition, showing that this bound transfers to the oracle groups under an explicit alignment condition. Extensive experiments show that prototype margins enrich shortcut-conflicting examples, classifier repair reshapes candidate evaluation, and ProME achieves the highest average worst-group accuracy among the compared methods with the same group-label access.

cs.LG↗

arXiv2026

Time-dependent multi-energy neutrino emission from symbiotic recurrent novae: the role of accretion disks

Symbiotic recurrent novae provide a unique laboratory for studying thermonuclear explosions, shock evolution, and nonthermal particle acceleration in dense circumstellar environments. In this work, we develop a time-dependent, multi-energy framework to describe neutrino emission from such systems, consistently incorporating both MeV neutrinos produced during thermonuclear runaway and GeV neutrinos generated through hadronic interactions in nova-driven shocks. Using RS Oph as a benchmark source, we model the evolution of the shock interacting with both the red giant wind and a dense accretion disk surrounding the white dwarf. We show that the resulting neutrino signal exhibits a characteristic two-component temporal structure: an early, rapidly rising MeV component tracing nuclear burning, followed by a delayed GeV component governed by shock propagation and particle acceleration. The presence of an accretion disk can significantly enhance the early-time GeV neutrino emission by providing a dense target for proton-proton interactions. This leads to a pronounced neutrino flux within the first few hours after eruption, a feature absent in wind-dominated scenarios. We further evaluate the detectability of these signals and find that while the MeV component remains below current detection thresholds, the GeV neutrino emission from nearby systems may become accessible to next-generation detectors. Our results highlight the critical role of the circumstellar structure in shaping nova neutrino emission and demonstrate that symbiotic recurrent novae are promising targets for future multi-messenger observations.

astro-ph.HE↗

arXiv2026

Conditionally Identifiable Latent-Environment Modeling for Out-of-Distribution Recommendation

Out-of-distribution (OOD) recommendation is vulnerable to preference shifts induced by a latent environment. Existing methods can infer latent states from logged interactions, yet the statistical meaning of the latent environment and its effect on preference remain underdetermined. We formulate this task as conditionally identifiable risk-aware recommendation (CI-RR) and propose Conditionally Identifiable Latent-Environment Recommendation (CILER). CILER uses a user-conditioned exponential family to model the latent environment and a feature-indexed polynomial to specify how it changes preference. It predicts by marginalizing item probabilities over the inferred environment distribution. Under sufficient variation, correct specification, and decoder regularity, CILER identifies the environment-sensitive representation up to the stated equivalence class. We further bound excess deployment log-risk by environment-inference error. Controlled studies test the observable consequences of sufficient variation and model specification. Experiments on three datasets show that CILER improves all twelve OOD ranking metrics under feature, temporal, and geographical shifts within shared support.

cs.IR↗

arXiv2026

Risk assessment of muon single-event effects for low-altitude aircraft

With the rapid development of low-altitude economy, the radiation environment safety of low-altitude aircraft such as drones and electric vertical take-off and landing aircraft has attracted increasing attention. Although the dense lower atmosphere traditionally serves as an effective shield against cosmic radiation, the shrinking feature sizes of modern integrated circuits greatly enhance their vulnerability to single-event effects (SEEs). This study quantitatively evaluates muon-induced SEE risks for low-altitude aircraft in various regions of China under both static cosmic-ray background and ground-level enhancement (GLE) events, aiming to provide critical guidance for the next-generation low-altitude aviation platforms.Using city-specific atmospheric models within the CORSIKA framework, we simulate atmospheric shower processes and obtain reliable energy spectra for low-energy muons (10-100 MeV). We also employ simulation data from other research groups to estimate muon-induced SEE cross sections for transistors at different process nodes, including bulk, FD-SOI, and FinFET technologies. By incorporating solar energetic particle spectra associated with GLE events, we assess muon-induced SEE risks under both static and GLE conditions. Our results show that under static conditions, flight control systems with 1 MB memory using advanced nodes below 45 nm and bulk transistors face non-negligible muon-induced SEE risks in all Chinese cities. In contrast, systems with FD-SOI transistors can effectively mitigate these risks. For large-memory systems (1 GB), redundancy or other hardening measures are essential regardless of the process technology. Regarding GLE events, we introduce the concept of muon hazard levels to evaluate regional risk variations. During GLEs, the increase in muon-induced SEE risk is negligible in mid-to-low latitude regions but becomes significant at high latitudes.

astro-ph.IM↗

arXiv2026

Photons in Media: A Second-Quantization Scheme Based on a Dirac-like Equation

We develop a second-quantization framework for photons based on the optical Dirac equation of source-free Maxwell theory in generic media. In this formulation, the electromagnetic field is recast as a four-component spinor-like wave function that admits both positive-energy and negative-energy solutions, which are naturally interpreted as photon and antiphoton states. By expanding the field in terms of single-photon eigenmodes, we construct a consistent quantization scheme in which the photon field operators obey bosonic commutation relations, in close analogy with the Dirac quantization of electrons. In structured media, the optical Dirac equation acquires effective mass and coupling terms induced by the dielectric tensor, analogous to an electronic Dirac-type structure. This allows photon propagation in media to be interpreted in terms of boosted spinor states and provides a unified description of vacuum and medium-modified dispersion relations. The framework further reveals a natural quantum-mechanical origin of transverse spin in structured electromagnetic fields, including evanescent waves, where spin components perpendicular to the propagation direction emerge from the underlying helicity structure. In the context of optical Dirac theory, this work presents a quantum field-theoretic description of photons in both vacuum and media, offering a new perspective on photon quantization, spin-orbit interaction, and light-matter coupling in structured optical systems.

physics.optics↗

arXiv2026

Data-driven modeling of Galactic diffuse emission with multi-wavelength observations

We present a data-driven investigation of Galactic diffuse emission. Using multi-frequency Planck radio/microwave maps (30-857 GHz) and Fermi-LAT gamma-ray data (50 MeV-814 GeV), we construct a nonlinear mapping between radio emission and gamma-ray intensity through supervised machine learning. Our models achieve high predictive accuracy (R^2 > 0.90 in the 0.1-10 GeV range), demonstrating that multi-frequency radio observations encode sufficient information to reconstruct both spatial morphology and spectral properties of diffuse gamma-ray emission. By analyzing model performance across different frequency bands and spatial regions, we identify high-frequency radio bands as the dominant predictor, providing direct empirical support for the hadronic origin of Galactic 0.1-10 GeV gamma rays, while low-frequency radio bands for the leptonic origin above 10 GeV. Residual maps reveal coherent large-scale structures, including Loop I and III, highlighting regions where standard interstellar emission models are incomplete or biased. Compared with the GALPROP model, our machine learning approach yields a higher R^2=0.95 and lower mean absolute relative error (14.7%) in the inner Galactic disk and the Galactic center region. Our results illustrate that machine learning serves as a physically interpretable tool for multi-messenger astrophysics, providing a data-driven baseline for separating non-standard emission components and deriving new constraints on cosmic-ray propagation and interstellar medium structure.

astro-ph.HE↗

arXiv2026

LHAASO J1849$-$0002: A Hybrid Lepto-Hadronic Interpretation of PeV Gamma-Ray Emission

Recently, LHAASO detected gamma-ray emission from the pulsar wind nebula (PWN) J1849-0001 extending up to approximately 2 PeV, providing strong evidence for PeV particle acceleration. To explain the origin of this ultra-high-energy emission, we investigate three physical scenarios: a pure leptonic model, a hadronic-dominated model, and a hybrid lepto-hadronic model. We show that while both pure leptonic and hadronic-dominated models can reproduce parts of the multiwavelength spectral energy distribution (SED), neither can simultaneously explain the entire dataset, particularly the PeV tail. The leptonic scenario requires an unrealistically high electron cutoff energy, while the hadronic model underpredicts the highest-energy emission. We therefore propose a hybrid model that combines inverse Compton emission from PWN electrons with hadronic interactions between escaped cosmic rays and a nearby molecular cloud. In this framework, a suppressed diffusion coefficient ($\sim 1\%$ of the Galactic average) is required to confine PeV particles in the source vicinity. This model successfully reproduces the full SED, including the approximately 2 PeV emission. We further calculate the associated neutrino flux, and show the sensitivity of NEON to this source. Our results support the interpretation that evolved PWNe embedded in complex environments can act as Galactic PeVatrons.

astro-ph.HE↗

arXiv2026

Collective Winds of Massive Star Clusters as the Dominant PeVatrons for Galactic Cosmic Rays

The knee feature in the cosmic-ray energy spectrum around 4 PeV is widely believed to have a Galactic origin, but the acceleration mechanism and identification of PeVatrons remain key open questions in high-energy astrophysics. Recent precise measurements by LHAASO reveal that the proton and helium spectra exhibit a common rigidity-dependent spectral break at ~ 3.5 PV, imposing a stringent constraint on source models. In this work, we construct, for the first time, a time-dependent cosmic-ray injection model that incorporates the full evolution of massive stars together with the dynamical development of wind termination shocks. We find that stellar winds of individual massive stars cannot explain the common spectral break observed by LHAASO, as they yield distinct rigidity cutoffs for protons and helium. By contrast, collective winds of massive star clusters naturally reconcile this discrepancy through the mixing effect of stars at different evolutionary stages. We propose a stellar-dominated model in which supernova remnants dominate the GeV-TeV range, individual stellar winds dominate the TeV range, and collective cluster winds dominate the PeV knee region. This model successfully reproduces the rigidity-dependent spectral features of various species near 100 GV and 0.1 PV. It further makes two testable predictions for future observations. Around 0.5 PV, the energy spectra of carbon and oxygen are expected to exhibit hardening similar to that of helium, which can be verified by LHAASO observations. In the multi-TV range, the energy spectrum of magnesium is not expected to show hardening similar to that observed for helium, carbon, and oxygen, which can be tested by DAMPE observations.

astro-ph.HE↗

arXiv2026

Radio Study of G76.9+1.0 Pulsar Wind Nebula

Pulsar Wind Nebulae (PWNe) are key astrophysical laboratories for high energy phenomena. Specifically, radio observations and related polarimetry are essential probes to understand acceleration and transport, as well as PWN interaction with environment. We aim to better study the multi-wavelength morphology and magnetic geometry of \gname\ PWN (a system between early and middle ages). We conduct high resolution VLA observations at 3 cm (X band), 6 cm (C band), and 13 cm (S band) and compare them with the archival Chandra X-ray data. We also performed spectral analysis and radio polarimetry based on our radio observations. Our new VLA observations reveal a north-south double-lobed PWN bracketing a bridge-like feature, with the pulsar clearly resolved at C and S bands. The polarization fraction reaches 30\% across all bands, with the bridge region showing ordered north-south magnetic fields aligned with the X-ray torus elongation, while the southern outer lobe exhibits fields not following such a direction and the northern lobe displays a more chaotic configuration. Notably, we detect a significant radio-X-ray anti-correlation near the pulsar, with bright radio emission appearing just beyond the compact X-ray PWN boundary, multiwavelength spectral analysis suggest distinct particle populations. The radio PWN spectral index steepens from $\alpha\sim-0.3$ in the inner bridge to $<-1.0$ in the outer lobes, yet we suggest it is less likely related to synchrotron cooling. We tried to use a thick torus model with toroidal $B$-field to reproduce observed features; the result implies possible particle deceleration in the radio PWN. The equipartition magnetic field strength is estimated to be $\sim$15.3\,$\mu$G.

astro-ph.HE↗

arXiv2026

Chasing the neutrino blazar candidates II: SED modeling with hadronic model

Blazars are promising candidates for high energy neutrino sources, yet the physical origin of their neutrino emission remains uncertain. In this work, we extend our previous study by modeling the broadband spectral energy distributions (SEDs) of 103 neutrino blazar candidates (NBCs) within a hadronic framework. To estimate the maximum possible neutrino output, we adopt an assumption in which the high energy emission is dominated by p gamma interactions and the contribution from leptonic inverse Compton scattering is strongly suppressed. From the SED modeling, we constrain nine key parameters describing the emission region and particle energy distributions. We perform a partial correlation analysis to investigate the relationship between neutrino luminosity and electromagnetic emission, and we found a weak or moderate correlation between optical R band and neutrino emission. Our model predicts prominent proton synchrotron emission peaking in the MeV band for most sources, with 99 out of 103 NBCs exhibiting proton synchrotron peaks within 0.1 to 100 MeV, highlighting the MeV band as a key window for distinguishing between leptonic and hadronic scenarios. Based on the model-predicted maximum neutrino fluxes, we find that three NBCs are potentially detectable by IceCube, while up to 22, 45, and 62 sources may be detectable by KM3NeT, NEON, and TRIDENT, respectively. These results provide testable predictions for future multi-messenger observations and offer new insights into the composition and radiation mechanisms of blazar jets.

astro-ph.HE↗

arXiv2026

Representation-Guided Parameter-Efficient LLM Unlearning

Large Language Models (LLMs) often memorize sensitive or harmful information, necessitating effective machine unlearning techniques. While existing parameter-efficient unlearning methods have shown promise, they still struggle with the forget-retain trade-off. This can be attributed to their reliance on parameter importance metrics to identify parameters that are important exclusively for the forget set, which is fundamentally limited by the superposition phenomenon. Due to the polysemantic nature of LLM parameters, such an importance metric may struggle to disentangle parameters associated with the forget and retain sets. In this work, we propose Representation-Guided Low-rank Unlearning (REGLU), a novel approach that leverages the geometric properties of representation spaces to achieve robust and precise unlearning. First, we develop a representation-guided initialization for LoRA that identifies the optimal subspace for selective forgetting. Second, we introduce a regularization loss that constrains the outputs of the LoRA update to lie in the orthogonal complement of the retain set's representation subspace, thereby minimizing interference with the model's performance on the retain set. We evaluate REGLU on the TOFU and WMDP benchmarks across multiple models. Our results demonstrate that REGLU consistently outperforms state-of-the-art baselines, achieving superior unlearning quality while maintaining higher model utility.

cs.CL↗

arXiv2026

Deciphering the IceCube Diffuse Neutrino Observations via AGN Variability

The physical origin of the diffuse neutrino background and its spectral break at $\sim$ 30 TeV remain a major puzzle in multi-messenger astrophysics. In this work, we demonstrate that this spectral feature is a natural consequence of AGN activity cycles and the resulting cosmic ray (CR) propagation. We present a unified model coupling the active and quiescent phases of AGNs, where CRs accelerated in the active core undergo subsequent diffusion and hadronic interactions in the host galaxy during the quiescent phase. The superposition of these distinct evolutionary phases yields dual spectral breaks, particularly the one at tens of TeV. Under realistic energetics, our model simultaneously accounts for the IceCube diffuse flux and fits the neutrino emissions of diverse sources, ranging from the blazar TXS 0506+056 to the Seyfert galaxies NGC 7469, CGCG 420-015, and the Circinus Galaxy. Our findings reveal that temporal variability is essential for deciphering the cosmic neutrino landscape and tracking high-energy CR escape.

astro-ph.HE↗

arXiv2026

Catching TeV emission from GRB 221009A and alike with LHAASO, LACT and SWGO

Gamma-Ray Bursts (GRBs) are the most energetic electromagnetic explosions in the universe. Recently, the Large High Altitude Air Shower Observatory (LHAASO) reported the breakthrough observation of GRB 221009A with gamma-ray energies beyond 13 TeV. This discovery, together with the previous GRB detection well above 100 GeV, confirms the production of very-high-energy (VHE, $\gtrsim 100$ GeV) radiation which might be a common component of all bright GRBs. It is reasonable to expect that bright GRBs are important targets for ground-based gamma-ray experiments. In this work, we estimate the detection rate for current and upcoming ground-based gamma-ray observatories including LHAASO, Large Array of Imaging Atmospheric Cherenkov Telescopes (LACT) and the Southern Wide-field Gamma-ray Observatory (SWGO) under two emission models with GRB~221009A as the template: first, that they all share the same intrinsic VHE spectral shape; second, they have the same environmental parameter and electron spectral index, governing their synchrotron self-Compton (SSC) emission. Using the long GRB luminosity and redshift distribution function obtained from the Fermi-GBM GRB samples, and accounting for the cosmological effects and extra-galactic background light (EBL) absorption, we derive the expected VHE flux at Earth. The sensitivity analysis for LHAASO, the upcoming LACT, and SWGO to evaluate their detection potential across specific redshift and luminosity ranges has been performed. The corresponding 5$\sigma$ detection rates of 221009A-like GRBs for the two emission models are: LHAASO, 0.04-0.05 yr$^{-1}$; LACT, 0.03-0.06 yr$^{-1}$; SWGO, 0.2-0.4 yr$^{-1}$. These rates can vary by up to $\approx 24\%$ due to different EBL models.

astro-ph.HE↗

arXiv2026

Radio Study of Vela X Cocoon

The evolution of pulsar Wind Nebulae (PWNe) influences how high energy particles in the vicinity are generated and transport. The Vela PWN (only $\sim300$\,pc away), provides a rather rare case between young and well-evolved systems. We therefore performed new 6 and 16\,cm high-resolution observations of the Vela X Cocoon region with the Australia Telescope Compact Array (ATCA). The observations reveal a complex region with a $\sim0.5^\circ$ major curved filament extending to far south from the pulsar, as well as other intersecting filaments and wisps. Our spectral analysis hints its connection with the PWN. Our results also found strongly linearly polarized emission, ordered and tangential $B$-field to the filaments. We find the rotation measure (RM) and polarization fraction (PF) along the filament are anti-correlated with the total intensity. We develop a simple 3D model of a spiral filament to explain these, while the PF distribution requires external interpretations such as interaction with the reverse shock. Comparison with archival data suggests that large scale features like the major filament are generally stable and large motions near the X-ray filament, all these confirm the distinction between radio and X-ray features.

astro-ph.HE↗