Searcharxiv⌕ Search

arXiv subjects

Bo Wang

Publications and source records attributed to Bo Wang.

At least 73 records · Page 4Linked to original sources
arXiv2026

Diffusion posterior sampling enables zero shot kilometre scale wind forecasting over complex terrain

Reliable prediction of near surface wind over complex terrain is limited by the mismatch between the spatial resolution of operational forecasts and terrain controlled local wind variability. Here, we develop KiloGen, a diffusion posterior sampling framework for kilometre scale wind forecast enhancement. KiloGen learns a high resolution vector wind prior from Weather Research and Forecasting (WRF) model simulations and constrains posterior sampling with 25 km forecasts from the European Centre for Medium Range Weather Forecasts (ECMWF) at inference time. This formulation avoids paired ECMWF and WRF training samples and an explicitly learned mapping from coarse to fine resolution. Applied over Shanxi, China, a region with complex mountainous terrain, KiloGen reconstructs terrain organized wind structures and restores high wavenumber variability while retaining the large scale evolution of the operational forecast. Station verification shows that KiloGen achieves the lowest overall wind speed root mean square error (RMSE) among the evaluated products, with larger benefits at elevated and topographically complex sites. The improvement is strongest under strong wind conditions, reducing RMSE by approximately 10% for observed winds above 20 m s^-1. Across 13 distinct strong wind events, KiloGen improves upon the 0.25 degree ECMWF forecast in all cases and outperforms the 0.1 degree ECMWF forecast in most cases. These results show that diffusion posterior sampling provides an effective approach for terrain aware kilometre scale wind forecast enhancement.

physics.ao-ph↗
arXiv2026

Chi-Square Wavelet Graph Neural Networks for Heterogeneous Graph Anomaly Detection

Graph Anomaly Detection (GAD) in heterogeneous networks presents unique challenges due to node and edge heterogeneity. Existing Graph Neural Network (GNN) methods primarily focus on homogeneous GAD and thus fail to address three key issues: (C1) Capturing abnormal signal and rich semantics across diverse meta-paths; (C2) Retaining high-frequency content in HIN dimension alignment; and (C3) Learning effectively from difficult anomaly samples with class imbalance. To overcome these, we propose ChiGAD, a spectral GNN framework based on a novel Chi-Square filter, inspired by the wavelet effectiveness in diverse domains. Specifically, ChiGAD consists of: (1) Multi-Graph Chi-Square Filter, which captures anomalous information via applying dedicated Chi-Square filters to each meta-path graph; (2) Interactive Meta-Graph Convolution, which aligns features while preserving high-frequency information and incorporates heterogeneous messages by a unified Chi-Square Filter; and (3) Contribution-Informed Cross-Entropy Loss, which prioritizes difficult anomalies to address class imbalance. Extensive experiments on public and industrial datasets show that ChiGAD outperforms state-of-the-art models on multiple metrics. Additionally, its homogeneous variant, ChiGNN, excels on seven GAD datasets, validating the effectiveness of Chi-Square filters. Our code is available at https://github.com/HsipingLi/ChiGAD.

cs.LG↗
arXiv2026

Optical observations on the young Type Ia SN 2021fxy with detached high velocity features

We present optical observations on the young type Ia supernova (SN Ia) SN 2021fxy obtained within a few days after the explosion, with a focus on its prominent high-velocity features (HVFs). It reached a $B$-band maximum of $M_{\rm max}(B) = -19.36\pm0.31$ mag, corresponding to a bolometric luminosity of $\sim 1.3\times10^{43}~\rm{erg~s^{-1}}$ with a synthesized $^{56}$Ni mass of $0.58\pm0.14$ M$_{\odot}$. The early spectra exhibit strong HVFs of intermediate-mass elements that are significantly detached from the photospheric components. In particular, the velocity of the Si II $\lambda6355$ HVFs follows a power-law evolution ($β\approx 0.1$), shallower than the expected photospheric velocity evolution expected for an assumed $n=10$ density profile ($β\approx 0.22$) under homologous expansion. This behavior is consistent with the HVFs forming in intrinsic ejecta structures at least partially decoupled from the bulk outer ejecta, providing a possible constraint on the explosion physics of SNe Ia.

astro-ph.HE↗
arXiv2026

Surrogate models for type II supernovae: Probing low-energy explosions and interaction-free regimes

To address the computational bottleneck of analyzing type II supernova samples from surveys such as the Legacy Survey of Space and Time, we present two STELLA-based neural-network surrogates: an interaction model for low-energy explosions with possible circumstellar-material (CSM) interaction and a photospheric model for standard interaction-free SNe IIP. Each uses an autoencoder to compress spectral energy distributions and an emulator to map physical parameters to the latent space. Latent-mixup regularization improves latent-space continuity, with ResNet blocks used for the interaction model and 2D CNNs for the photospheric model. Their normalized test-set reconstruction MSEs are approximately 9.1e-5 and 1.0e-4, respectively. Applied to SN 2005cs, the interaction model favors a low-mass progenitor, M_ZAMS = 10.40(+0.04/-0.05) M_sun, and confined dense CSM, providing a scenario consistent with direct imaging and helping resolve the historical mass discrepancy. For SN 2012aw, it recovers M_ZAMS = 11.05(+0.06/-0.06) M_sun, consistent with previous studies. For SN 1999em, the photospheric model gives M_ZAMS = 10.05(+0.07/-0.04) M_sun, broadly consistent with preexplosion imaging limits without explicit CSM modeling. These surrogates reduce full Bayesian inference from days to minutes and enable rapid physical characterization of large supernova samples.

astro-ph.SR↗
arXiv2026

Analyzing Functional Data with a Mixture of Covariance Structures Using a Curve-Based Sampling Scheme

Motivated by distinct walking patterns in real-world free-living gait data, this paper proposes an innovative curve-based sampling scheme for the analysis of functional data characterized by a mixture of covariance structures. Traditional approaches often fail to adequately capture inherent complexities arising from heterogeneous covariance patterns across distinct subsets of the data. We introduce a unified Bayesian framework that integrates a nonlinear regression function with a continuous-time hidden Markov model, enabling the identification and utilization of varying covariance structures. One of the key contributions is the development of a computationally efficient curve-based sampling scheme for hidden state estimation, addressing the sampling complexities associated with high-dimensional, conditionally dependent data. This paper details the Bayesian inference procedure, examines the asymptotic properties to ensure the structural consistency of the model, and demonstrates its effectiveness through simulated and real-world examples.

stat.ME↗
arXiv2026

X-rays breaking out of pre-explosion ejecta mark a supernova's first light

Massive stars die as core-collapse supernovae, whose optical light emerges days after the implosion. Theory predicts that the initial collapse-driven shock, upon breaking through the star and dense circumstellar medium, emits a brief thermal flash of soft X-rays and ultraviolet. Yet these elusive first signals have remained largely undetected, owing to limited wide-field soft X-ray monitoring. Here we report the discovery of a soft X-ray flash, EP260321a, followed days later by a broad-lined supernova from an envelope-stripped progenitor. Its X-ray spectrum, best modeled with blackbody, establishes it as the long-sought archetypal shock breakout. The burst's duration and energetics place the breakout at a radius of 300 solar radii, tracing a dense surrounding shell and revealing abrupt mass ejection within the final month before collapse.

astro-ph.HE↗
arXiv2026

Finite Boundary-Layer Residence Certificates for Non-Strict Control Barrier Functions

Non-strict control barrier function (CBF) conditions guarantee safety through forward invariance, but they do not preclude trajectories from remaining near the safe-set boundary for extended continuous time intervals. This paper develops a finite boundary-layer residence certificate for such settings. The certificate preserves the standard non-strict CBF safety condition and uses a bounded auxiliary function whose derivative is bounded away from zero in a prescribed boundary layer, yielding an explicit upper bound on every uninterrupted residence interval. For control-affine systems, a selected auxiliary branch is implemented as an additional affine constraint in a CBF-QP, and a tangential-input compatibility condition is given to ensure simultaneous feasibility with the hard CBF constraint for unconstrained inputs. A local-chart version handles angular or multi-valued auxiliary functions such as $\operatorname{atan2}$. Single-integrator, double-integrator, and nonholonomic unicycle examples illustrate the resulting radial--tangential construction and its local-chart and feasibility limitations.

eess.SY↗
arXiv2026

Classifying the hidden-charm pentaquarks via a flavor mixing scheme

In this work, we propose a scheme to classify the molecular states consisting of ground single-charm baryons ($Λ_c$, $Ξ_c$, $Σ_c^{(*)}$, $Ξ_c^{\prime(*)}$, $Ω_c^{(*)}$) and $\bar{D}^{(*)}/\bar{D}_s^{(*)}$ mesons. Within this framework, all considered baryon-meson systems are categorized according to the flavor components of their light degrees of freedom. We briefly illustrate how this classification scheme can consistently explain the experimentally observed $P_c$ and $P_{cs}$ states. This framework also predicts the existences of single-strange and double-strange hidden-charm bound states. The attractive interactions of these states arise from channel mixing between $Σ_c^{(*)}\bar{D}_s^{(*)}$ and $Ξ_c^{\prime(*)}\bar{D}^{(*)}$ for single-strange systems, and mixing between $Ξ_c^{\prime(*)}\bar{D}_s^{(*)}$ and $Ω_c^{(*)}\bar{D}^{(*)}$ for double-strange systems, respectively. Using parameters fitted from the measured $P_c$ and $P_{cs}$ states, we systematically present the predicted mass spectra for these single- and double-strange hidden-charm bound states.

hep-ph↗
arXiv2026

Token-Based Affordance Grounding with Large Vision-Language Models

Affordance grounding aims to localize image regions that support a specific action, serving as a core capability for physical intelligence and embodied perception. Previous studies have primarily relied on weakly supervised learning with action labels from exocentric images. However, these methods often struggle with visually ambiguous exocentric images containing co-occurring actions; moreover, they fail to distinguish semantically similar actions because existing methods typically rely on brief action phrases that lack rich semantic details for action-specific localization. Although large vision-language models (LVLMs) encode rich action semantics and their action-conditioned textual outputs implicitly contain spatial cues, they do not directly provide action-specific spatial localization. To address these problems, we propose TokAG, a zero-shot affordance grounding framework that exploits the token-level semantic-spatial signals in LVLMs to localize action-relevant regions without external supervision. We observe that attention maps associated with different LVLM output tokens vary significantly, with many attending to irrelevant regions such as the background. Thus, we introduce a spatial-aware token-selection mechanism to systematically evaluate each output token and select the one whose attention maps exhibit dominant activation over the target object, instead of relying on arbitrary attention maps. By extracting these object-focused attention maps, we transform the LVLM's implicit semantic signals into zero-shot affordance heatmaps. Our zero-shot framework consistently outperforms prior weakly supervised approaches across multiple benchmarks, improving NSS by 10.7% on the unseen split of AGD20K and by 29.7% on HICO-IIF. The code and models will be made publicly available.

cs.CV↗
arXiv2026

Nonholonomic Source Seeking by Torque Tuning: Local and Semi-Global Feedbacks

This paper studies source seeking for a torque-controlled nonholonomic vehicle with a laterally displaced scalar sensor. The vehicle has constant forward speed, while its yaw motion is controlled by torque input with unknown inertia and damping. The objective is to steer the vehicle to a source-centered circular motion so that the lateral sensor approaches the unknown source, without using position, heading, source-location, gradient, or source-value information. The proposed torque law combines a fast oscillatory component, which generates averaged steering through symmetric-product approximation, with a slowly tuned bias component, which selects the desired orbit. Two bias-tuning designs are developed. The first is an output-feedback design using only the scalar measurement; it applies a Lie-bracket extremum-seeking update and yields local practical stability. The second is a velocity-assisted design using forward-speed and yaw-rate measurements; it tunes the bias through the yaw-rate tracking error and yields a globally asymptotically stable averaged system, implying semi-global practical stability of the original system. Simulations illustrate the proposed designs.

eess.SY↗
arXiv2026

Terahertz oscillation of $180^{\circ}$ domain walls in ferroelectric membranes

A fundamentally intriguing yet not well understood topic in the field of ferroelectrics is the collective excitation of domain walls (DWs), with potential applications to DW-based nanoelectronic and optoelectronic devices. Here we use dynamical phase-field simulations to identify the collective modes of an Ising-type $180^{\circ}$ DW in a uniaxially strained BaTiO3 membrane. The membrane, which concurrently functions as a cavity for polarization and acoustic waves, permits cavity-enhanced resonant excitation of polarization waves. The simulation reveals an unconventional DW sliding mode that exhibits a bulk-polarization-charge-driven nonzero resonant frequency in the sub-terahertz (THz) regime with a dynamically changing internal structure during sliding. These features differ from the previously reported zero-frequency DW sliding mode or the surface-polarization-charge-driven DW sliding mode. The effect of strain on the frequency of this unconventional DW sliding mode and other previously known THz DW eigenmodes is investigated by dynamical phase-field simulations and interpreted by eigenmode analysis or analytical calculation in a simplified one-dimensional (1D) system. These results provide new insights into the high-frequency dynamics of ferroelectric DWs and suggest opportunities for realizing strain control of phonon-DW resonance, and more broadly, for discovering and controlling unconventional DW modes in conventional domain patterns with applications to reconfigurable THz and optical devices.

cond-mat.mtrl-sci↗
arXiv2026

Knowledge-Enhanced Agentic Vulnerability Repair

Frontier foundation models have changed the math on vulnerability discovery, but the bigger challenge is how the remediation side keeps up. Despite recent progresses in Automated Vulnerability Repair (AVR), current solutions struggle to reliably identify the root causes of vulnerabilities, and insufficiently utilize the prior fix knowledge to guide the patch generation process, thus undermining their effectiveness in practice. To address this gap, we propose KeaRepair, a novel agentic AVR approach that grounds patch generation in verified program facts and high-level vulnerability knowledge. Specifically, KeaRepair first extracts multi-dimensional vulnerability knowledge from historical vulnerability-patch pairs from dual complementary views, and constructs dedicated retrieval knowledge bases. It then employs a tool-augmented agent that performs ReAct-style reasoning to collect verified program facts for vulnerability diagnosis. Finally, based on the diagnostic results, KeaRepair performs knowledge-level retrieval-augmented patch generation and iteratively refines patches through a closed-loop validation process involving compilation, PoC replay, and test-suite execution. Experimental results show that KeaRepair significantly outperforms existing AVR approaches on 55 reproducible C/C++ vulnerabilities. When paired with Gemini-3.1-Pro, KeaRepair successfully repairs 46 vulnerabilities, achieving a repair rate of 83.64%. Moreover, KeaRepair fixes six unique vulnerabilities that none of the baselines can address, and further demonstrates strong cross-language generalizability.

cs.SE↗
arXiv2026

Regime-Aware Peer Specialization for Robust RAG under Heterogeneous Knowledge Conflicts

Retrieval-augmented generation (RAG) improves language models by grounding generation in external context. However, it can be fragile when the retrieved context conflicts with the model's parametric knowledge. Such conflicts span a reliability spectrum, ranging from reliable and partially reliable evidence to adversarial context. Existing remedies often handle such heterogeneous conflicts with regime-agnostic supervision, which can conflate incompatible learning signals across reliability regimes. To disentangle these signals, we propose RAPS-DA, a regime-aware peer specialization framework that addresses conflict at two complementary granularities. At the sample level, conflicts are divided into three regimes, including Grounding, Arbitration, and Resistance, with one same-scale peer specialist trained per regime from a shared base model. Each sample is then hard-routed to its regime-matched peer for on-policy reverse-KL supervision. At the token level, a dual-layer selector uses inter-teacher disagreement, student-teacher divergence, and student entropy to filter uninformative or unstable tokens, upweight confidently misaligned ones, and gradually focus supervision on high-conflict tokens as the student matures. Gains stem from specialization at a fixed model scale, not from a stronger teacher, and the peer specialists exist only during training, so the deployed student requires no regime labels or peer access. Experiments on five conflict scenarios and two out-of-distribution benchmarks show RAPS-DA surpasses all prompting, decoding, fine-tuning, RL, and single-teacher baselines.

cs.CL↗
arXiv2026

Elemental Abundances from Off-center Carbon Burning in Accreting CO White Dwarfs: Implications for SN 2021yfj-like events

SN 2021yfj is a recently discovered interacting supernova that exhibits narrow emission lines of Si, S, and Ar, indicating the presence of circumstellar material (CSM) enriched with these elements surrounding the progenitor prior to the explosion. The origin of SN 2021yfj-like events remains uncertain. Recent work proposed that the SN 2021yfj-like events may stem from the double WD merger scenario, in which the merger of a Si-rich WD with a more massive ONe WD tidally strips about 0.3Msun of Si-rich material to form the CSM. If the merger subsequently triggers a supernova explosion, the interaction between the ejecta and the CSM can reproduce the observed light curve of SN 2021yfj. In this scenario, the progenitor system is a CO WD + He star binary, in which the CO WD accretes He-rich material from the He star. The accumulated material can trigger off-center carbon burning, potentially leading to the formation of a Si-rich WD. However, it remains unclear whether such off-center carbon burning can produce Si, S, and Ar in amounts comparable to those inferred for the CSM of SN 2021yfj. In this work, we simulate the evolution of a CO WD accreting He-rich material using time-dependent mass-accretion rates. Our results show that off-center carbon burning in the accreting CO WD can produce significant amounts of Si and S. We further found that the resulting elemental abundances are strongly affected by the initial carbon abundance of the WD. Based on our results, we suggest that the double WD merger scenario may provide a viable progenitor channel for SN 2021yfj-like events.

astro-ph.SR↗
arXiv2026

Mostly Automatic Translation of Language Interpreters from C to Safe Rust

Translating C programs to safe Rust is challenging owing to significant differences in typing constraints, ownership, and borrowing rules. Interpreter programs are particularly important targets for such translation, as they often handle untrusted inputs and suffer from memory-related vulnerabilities. We present Reboot, a mostly-automatic technique that translates real-world interpreter programs from C to safe Rust. Using Reboot, we have translated six interpreters ranging from 6k to 23k lines of C code to safe Rust, with each translation requiring only 1 to 11 brief user interventions. All translations pass 100% of the provided test suites, and achieve 62%--92% pass rates on separately created validation tests that were never exposed to the system. A security case study on mujs shows that memory vulnerabilities such as heap buffer overflows and use-after-free present in C are eliminated in the safe Rust translation. Two ideas underpin Reboot. First, feature reduction decomposes the translation by program features, creating a sequence of milestones where each is a complete, testable program; the translation starts from the simplest version and incrementally restores features, with each milestone validated before proceeding. Second, a multi-agent architecture orchestrates inherently unreliable coding agents through automated validation and feedback, keeping long-running translation workflows on track with minimal human involvement. An ablation study confirms that feature reduction improves translation correctness compared to using multi-agent translation alone, with 6%--20% improvements in pass rates on validation test suites.

cs.PL↗
arXiv2026

Metric results for dyadic approximation on the middle-third Cantor set

Let $C$ be the middle-third Cantor set and $μ$ be the Cantor-Lebesgue measure on $C$. A conjecture of Velani states that $μ(W_2(τ))=0$ if $τ>1$ and $μ(W_2(τ))=1$ if $0<τ\leq 1$, where $W_2(τ)=\left\{x\in[0,1]: \|2^nx\| \frac{1}γ-\frac{1-γ}{3-γ}\,(\approx 1.429)$ and $0<τ<\fracγ{12}\,(\approx 0.052)$, where $γ=\frac{\log2}{\log3}$ is the Hausdorff dimension of $C$. This improves the known results on both the null part ($τ>\frac{1}γ-\frac{0.078(1-γ)}{γ(2-γ)}\approx 1.552$, due to Allen, Baker, Chow, and Yu (2023)) and the full measure part ($0<τ\leq 0.01$, due to Baker (2025)). Our key innovation is to establish the estimate \[\sum_{n=1}^{N}|\widehatμ(h2^n)|^2\ll N^{1-γ}\] and its consequences: \[ \sum_{n=1}^{N}|\widehatμ(h2^n)|\ll N^{1-\fracγ{2}},\quad \sum_{n=1}^{N}n^{- σ}|\widehatμ(h2^n)|\ll_σ N^{1-\fracγ{2}-σ},\] where $0<σ<1-\fracγ{2}$, and all estimates are uniform in $h\in\mathbb{Z}\setminus\{0\}$. For the full measure part, our approach also generalizes to self-similar measures on a class of missing-digit sets.

math.NT↗
arXiv2026

Can dwarf novae produce type Ia supernovae?

Accreting white dwarfs (WDs) are considered one of the most promising progenitor candidates for Type Ia supernovae (SNe Ia). Dwarf novae (DNe), a subclass of cataclysmic variables (CVs), consist of a carbon--oxygen (CO) WD accretor and a low-mass donor star, which may be either a main-sequence (MS) star or a slightly evolved subgiant. Previous studies have suggested that, under the thermal--viscous disk instability mechanism, the time-averaged accretion rate in long-period DNe may approach the regime of stable hydrogen burning, potentially allowing the WD to grow toward the Chandrasekhar mass, (M_{\rm Ch}). However, whether such periodic accretion can sustain stable hydrogen burning and lead to WD mass growth remains uncertain. In this work, we explore whether high accretion rates on short periodic timescales can maintain stable hydrogen burning on the WD surface and drive the WD toward (M_{\rm Ch}). Using Modules for Experiments in Stellar Astrophysics (MESA), we investigate the mass-retention efficiency of WDs undergoing intermittent DN-like accretion and examine its dependence on duty cycle, accretion rate, and initial WD properties. We find that periodic accretion fails to maintain stable hydrogen burning. During quiescent phases, the WD cools and becomes increasingly degenerate, leading to nova outbursts with progressively decreasing mass-retention efficiency and ultimately preventing further WD mass growth. We therefore suggest that DNe are unlikely to be progenitors of SNe Ia.

astro-ph.SR↗
arXiv2026

Recommendation as Generation: Unifying Personalized Video Generation and Recommendation at Industrial Scale

Traditional short-video recommendation systems match user interest to a fixed pool of pre-produced videos, which limits their ability to capture fine-grained and dynamic preferences. We propose Recommendation-as-Generation (RaG), a new paradigm that generates personalized videos on demand from inferred user interest. Our framework unifies generative recommendation and video generation through shared semantic IDs (SIDs), which disentangle video representation into content semantics and creative style semantics, enabling both fine-grained modeling of user interest and controllable generation of interest-aligned videos. We further develop Video Generation Agents (VGAs) that are conditioned on inferred SIDs to drive hierarchical planning and refinement for video creation, including visual composition, audio alignment, and artistic effect enhancement. To optimize the framework, we effectively introduce a synergistic cross-domain reward learning mechanism that jointly enforces interest alignment, user feedback, and video quality assessment. We deploy RaG on an industrial-scale platform with over 400 million daily active users and evaluate it in a revenue-critical advertising scenario. Online A/B tests show up to 1.87% ad revenue improvement compared to a strong production GRM baseline, demonstrating its effectiveness in driving further revenue gains beyond generative recommendation. Our results highlight a closed-loop generative system as a promising paradigm for integrating personalized video generation into recommendation.

cs.IR↗