arXiv subjects

Zhiwei Hu

Publications and source records attributed to Zhiwei Hu.

At least 19 recordsLinked to original sources

arXiv2026

QoEReasoner: An Agentic Reasoning Framework for Automated and Explainable QoE Diagnosis in RANs

Diagnosing Quality-of-Experience (QoE) degradations in operational Radio Access Networks (RANs) is a critical but notoriously complex task, traditionally requiring labor-intensive expert analysis over high-dimensional, cross-layer telemetry. While Large Language Models (LLMs) offer unprecedented reasoning capabilities, they are fundamentally unsuited for raw RANs troubleshooting: they fail at numeric time-series analysis, hallucinate protocol-violating causal links, and lack the stateful rigor required for multi-step fault localization. To bridge this gap, we present QoEReasoner, an end-to-end, LLM-driven agentic system designed for automated and explainable QoE diagnosis. QoEReasoner tames the inherent unpredictability of LLMs by grounding their reasoning in the physical realities of the network. It employs deterministic tools to reliably translate raw numeric KPIs into structured evidence, enforces protocol-consistent fault propagation through a domain-specific Knowledge Base, and leverages a Historical Bank of expert-validated cases to guide hypothesis generation. A stateful central planner orchestrates this closed-loop process across anomaly detection, causal tracing, and root-cause localization. Evaluations on real-world operational RANs datasets demonstrate that QoEReasoner outperforms strong baselines by 18\%-40\% in accuracy across multiple diagnostic tasks. Furthermore, it reduces diagnostic time from approximately 30 minutes of manual expert analysis to just 3 minutes per session, delivering highly interpretable, expert-grade reports while remaining robust across diverse LLM backbones.

cs.MA↗

arXiv2026

Enhancing collective spin squeezing via one-axis twisting echo control of individual atoms

Spin squeezing generated via inter-atom entanglement in multilevel atomic ensembles provides a powerful resource for quantum-enhanced metrology. Existing schemes that harness internal atomic degrees of freedom to boost squeezing typically encode the collective squeezing in complex superpositions of magnetic sublevels, which complicates state control and limits practical applications. Here, we propose a coherent control scheme that simultaneously enhances collective spin squeezing and maps the resulting atom-atom entanglement onto two well-defined magnetic sublevels suitable for subsequent metrology experiments. Our protocol sandwiches a quantum non-demolition measurement between two internal one-axis-twisting interactions arranged in an echo sequence. We show that this approach can optimally leverage internal states to boost the inter-atom entanglement and, at the same time, encode it in two magnetic sublevels, which is readily convertible into metrologically useful spin squeezing. Our results offer a straightforward and efficient strategy for generating highly entangled yet readily accessible quantum states in multilevel atomic systems.

quant-ph↗

arXiv2025

Extreme polaritonic interactions in a room-temperature deterministic sub-nanocavity quantum electrodynamic platform

Pushing nanoscale optical confinement to its ultimate limits defines the regime of nano-cavity quantum electrodynamics (nano-cQED), where light--matter interactions approach the fundamental quantum limits of individual atoms, e.g., picocavities. However, realizing such extreme confinement in a stable and controllable manner remains a key challenge. Here, we introduce a van der Waals material-based nano-cQED platform by coupling monolayer MoS2 excitons to plasmonic sub-nanocavities formed via assembly of ultrasmall gold clusters (3-5 nm) in the nanogap of a nanoparticle-on-mirror nanocavity. These clusters emulate the field-confining role of atomic protrusions of the picocavities through a resonance-insensitive lightning-rod effect, achieving deep-subwavelength mode volumes. In this nano-cQED testbed, we observe pronounced multi-branch Rabi splittings and ultrastrong lower-branch polaritonic photoluminescence with up to 10^4-fold enhancement. This deterministic architecture provides a controllable pathway to access picocavity-like behavior and opens new opportunities for single-molecule spectroscopy and the exploration of nano-cQED.

physics.optics↗

arXiv2025

Plan Then Retrieve: Reinforcement Learning-Guided Complex Reasoning over Knowledge Graphs

Knowledge Graph Question Answering aims to answer natural language questions by reasoning over structured knowledge graphs. While large language models have advanced KGQA through their strong reasoning capabilities, existing methods continue to struggle to fully exploit both the rich knowledge encoded in KGs and the reasoning capabilities of LLMs, particularly in complex scenarios. They often assume complete KG coverage and lack mechanisms to judge when external information is needed, and their reasoning remains locally myopic, failing to maintain coherent multi-step planning, leading to reasoning failures even when relevant knowledge exists. We propose Graph-RFT, a novel two-stage reinforcement fine-tuning KGQA framework with a 'plan-KGsearch-and-Websearch-during-think' paradigm, that enables LLMs to perform autonomous planning and adaptive retrieval scheduling across KG and web sources under incomplete knowledge conditions. Graph-RFT introduces a chain-of-thought fine-tuning method with a customized plan-retrieval dataset activates structured reasoning and resolves the GRPO cold-start problem. It then introduces a novel plan-retrieval guided reinforcement learning process integrates explicit planning and retrieval actions with a multi-reward design, enabling coverage-aware retrieval scheduling. It employs a Cartesian-inspired planning module to decompose complex questions into ordered subquestions, and logical expression to guide tool invocation for globally consistent multi-step reasoning. This reasoning retrieval process is optimized with a multi-reward combining outcome and retrieval specific signals, enabling the model to learn when and how to combine KG and web retrieval effectively.

cs.AI↗

arXiv2025

Spatiotemporal Raman Probing of Molecular Transport in sub-2-nm Plasmonic Quasi-2D Nanochannels

Capturing molecular dynamics in nanoconfined channels with high spatiotemporal resolution is a key challenge in nanoscience, crucial for advancing catalysis, energy conversion, and molecular sensing. Bottom-up ultrathin plasmonic nanogaps, such as nanoparticle-on-mirror (NPoM) structures, are ideal for ultrasensitive probing due to their extreme light confinement, but their perceived sealed geometry has cast doubt on the existence of accessible transport pathways. Here, counterintuitively, we demonstrate that ubiquitous ligand-capped NPoM-type nanogaps can form a natural quasi-two-dimensional nanochannel, supporting molecular transport over unprecedented length scales ($\gtrsim5$ $\mu$m) with an extreme aspect ratio ($>10^3$). Using wavelength-multiplexed Raman spectroscopy, we resolve the underlying centripetal infiltration pathway with a spatial resolving power of $\sim$20 nm. This redefines the NPoM architecture as a sensitive, \textit{in-situ}, all-in-one "transport-and-probe" platform, enabling real-time, reusable monitoring of analyte with $\sim$10$^{-11}$ M. This work establishes a versatile new platform for advancing super-resolved \textit{in-situ} molecular sensing, nanoscale physicochemical studies, and on-chip nanophotofluidics.

physics.optics↗

arXiv2025

Quantitative Benchmarking of Remote Excitation in Plasmonic Sensing with Enhanced Signal-to-Noise Ratio

Remote excitation using guided optical modes -- such as waveguides, fibers, or surface waves -- offers a promising alternative to direct optical excitation for surface-enhanced Raman scattering (SERS), particularly in applications requiring reduced heating, minimal invasiveness, and on-chip integration. However, despite its widespread use, systematic comparisons between remote and direct excitation remain limited. Here, we quantitatively benchmark both schemes by measuring power-dependent SERS responses from individual plasmonic nanogaps. We statistically analyze the maximum achievable SERS intensity before structural degradation, extract local temperatures, and evaluate signal-to-noise ratios (SNR). Our findings reveal that both remote and direct SERS share a common electric-field limit, despite exhibiting different levels of heating. This suggests that spectral evolution is primarily governed by the local electric field, which drives nanoscale atomic migration rather than excessive heating. Nonetheless, the lower heating associated with remote excitation enhances the Raman SNR by approximately 30%, improving measurement quality without compromising signal strength. This study establishes a quantitative framework for evaluating excitation strategies in plasmonic sensing, and challenges common assumptions about the role of heating in nanostructural stability under strong optical excitation.

physics.optics↗

arXiv2025

Multi-level Mixture of Experts for Multimodal Entity Linking

Multimodal Entity Linking (MEL) aims to link ambiguous mentions within multimodal contexts to associated entities in a multimodal knowledge base. Existing approaches to MEL introduce multimodal interaction and fusion mechanisms to bridge the modality gap and enable multi-grained semantic matching. However, they do not address two important problems: (i) mention ambiguity, i.e., the lack of semantic content caused by the brevity and omission of key information in the mention's textual context; (ii) dynamic selection of modal content, i.e., to dynamically distinguish the importance of different parts of modal information. To mitigate these issues, we propose a Multi-level Mixture of Experts (MMoE) model for MEL. MMoE has four components: (i) the description-aware mention enhancement module leverages large language models to identify the WikiData descriptions that best match a mention, considering the mention's textual context; (ii) the multimodal feature extraction module adopts multimodal feature encoders to obtain textual and visual embeddings for both mentions and entities; (iii)-(iv) the intra-level mixture of experts and inter-level mixture of experts modules apply a switch mixture of experts mechanism to dynamically and adaptively select features from relevant regions of information. Extensive experiments demonstrate the outstanding performance of MMoE compared to the state-of-the-art. MMoE's code is available at: https://github.com/zhiweihu1103/MEL-MMoE.

cs.CV↗

arXiv2025

Plasmonic Nanoparticle-in-nanoslit Antenna as Independently Tunable Dual-Resonant Systems for Efficient Frequency Upconversion

Dual-band plasmonic nanoantennas, exhibiting two widely separated user-defined resonances, are fundamental building blocks for the investigation and optimization of plasmon-enhanced optical phenomena, including photoluminescence, Raman scattering, and various nonlinear effects such as harmonic generation or sum-frequency generation, parametric down-conversion, etc. The nanoparticle-on-slit (NPoS) or nanoparticle-in-groove (NPiG) is a recently proposed dual-band antenna with independently tunable resonances at mid-infrared and visible wavelengths. It was used to enhance the corresponding sum- and difference-frequency generation processes from optimally located molecules by an estimated $10^{13}$-fold. However, the theoretical understanding of such structures and their eigenmodes remains poor, hindering further optimization and limiting broader applications. Here, we explore a diverse range of nanocavity-like quasi-normal modes (QNMs) supported by NPoS structures, examining the contributions of both their near-field (i.e., giant photonic density of states) and far-field (i.e., spatial radiation patterns) characteristics to frequency upconversion. We identify methods for independently tuning the visible and mid-infrared resonances while conserving a good mode overlap in the near field, which is essential for efficient nonlinear processes. Moreover, through mode analysis, we unveil an experimentally unexplored fundamental resonance with greater field enhancement and much-improved mode overlap with the mid-infrared field, which could, in principle, further boost the mid-infrared upconversion efficiency by 5-fold compared to existing results. This work helps to rationalize and optimize the enhancement of nonlinear effects across a wide spectral range using a flexible and experimentally attractive nanoplasmonic platform.

physics.optics↗

arXiv2025

Seeing Beyond RGB Capabilities: Data-Driven and Physics-Guided Broadband Spectral Extrapolation of Plasmonic Nanostructures by Deep Learning

Localized surface plasmons can confine light within a deep-subwavelength volume comparable to the scale of atoms and molecules, enabling ultrasensitive responses to near-field variations. On the other hand, this extreme localization also inevitably amplifies the unwanted noise from the response of local morphological imperfections, leading to complex spectral variations and reduced consistency across the plasmonic nanostructures. Seeking uniform optical responses has therefore long been a sought-after goal in nanoplasmonics. However, conventional probing techniques by dark-field (DF) confocal microscopy, such as image analysis or spectral measurements, can be inaccurate and time-consuming, respectively. Here, we introduce SPARX, a deep-learning-powered paradigm that surpasses conventional imaging and spectroscopic capabilities. In particular, SPARX can batch-predict broadband DF spectra (e.g., 500-1000 nm) of numerous nanoparticles simultaneously from an information-limited RGB image (i.e., below 700 nm). It achieves this extrapolative inference beyond the camera's capture capabilities by learning the underlying physical relationships among multiple orders of optical resonances. The spectral predictions only take milliseconds, achieving a speedup of three to four orders of magnitude compared to traditional spectral acquisition, which may take from hours to days. As a proof-of-principle demonstration for screening identical resonances, the selection accuracy achieved by SPARX is comparable to that of conventional spectroscopy techniques. This breakthrough paves the way for consistent plasmonic applications and next-generation microscopies.

physics.optics↗

arXiv2025

High Quality Single Crystal of Kitaev Spin Liquid Candidate Material RuBr3 Synthesized under High Pressure

Kitaev quantum spin liquids have attracted significant attention in condensed matter physics over the past decade. To understand their emergent quantum phenomena, high-quality single crystals of substantial size are essential. Here, we report the synthesis of single crystals of the Kitaev quantum spin liquid candidate RuBr3, achieving millimeter-sized crystals through a self-flux method under high pressure and high temperature conditions. The crystals exhibit well-defined cleavage planes with a lustrous appearance. Transport characterizations exhibit a narrow band-gap semiconducting behavior with 0.13 eV and 0.11 eV band-gap in ab plane and along c axis, respectively. Magnetic measurement shows a transition to antiferromagnetic (AFM) state at approximately 29 K both in ab plane and along c axis. Notably, the N\'eel temperature increases to 34 K with an applied magnetic field of up to 7 T in the ab plane, but without any change along c axis. The large size and high quality of RuBr3 single crystals provide a valuable platform for investigating various interactions, particularly the Kitaev interaction, and for elucidating the intrinsic physical properties of Kitaev quantum spin liquids.

cond-mat.str-el↗

arXiv2024

Local electronic properties of La3Ni2O7 under pressure

The recent discovery of superconductivity in $\rm La_3Ni_2O_7$ has attracted significant attention due to its high critical temperature and analogy to cuprate oxides. The oxidation and spin states of Ni ions are among the most important local properties in this compound, extensively discussed in the context of its superconductivity. Despite their direct link to the electron filling configurations of the relevant $\rm 3d_{x^2-y^2}$ and $\rm 3d_{z^2}$ orbitals, these local electronic properties of $\rm La_3Ni_2O_7$ yet to be systematically investigated. In this work, we address this issue using x-ray absorption spectroscopy (XAS) and x-ray emission spectroscopy (XES) measurements under pressure. Comparison of Ni \textit{K}-edge XAS and $\rm K\beta$ XES with the reference spectra of $\rm NiO$ and $\rm LaNiO_3$ shows that Ni ions, with an average valence of $\sim 2.53+$, are in a low-spin ($\rm S = 1/2$) ground state under ambient conditions. High pressure XAS and XES data clearly show that the oxidation ($\sim 2.5+$) and spin ($\rm S = 1/2$) states of Ni ions remain stable across the investigated pressure (up to 30 GPa) and temperature (down to 10 K) ranges, ruling out previously proposed spin transition scenarios.

cond-mat.supr-con↗

arXiv2024

Multi-level Matching Network for Multimodal Entity Linking

Multimodal entity linking (MEL) aims to link ambiguous mentions within multimodal contexts to corresponding entities in a multimodal knowledge base. Most existing approaches to MEL are based on representation learning or vision-and-language pre-training mechanisms for exploring the complementary effect among multiple modalities. However, these methods suffer from two limitations. On the one hand, they overlook the possibility of considering negative samples from the same modality. On the other hand, they lack mechanisms to capture bidirectional cross-modal interaction. To address these issues, we propose a Multi-level Matching network for Multimodal Entity Linking (M3EL). Specifically, M3EL is composed of three different modules: (i) a Multimodal Feature Extraction module, which extracts modality-specific representations with a multimodal encoder and introduces an intra-modal contrastive learning sub-module to obtain better discriminative embeddings based on uni-modal differences; (ii) an Intra-modal Matching Network module, which contains two levels of matching granularity: Coarse-grained Global-to-Global and Fine-grained Global-to-Local, to achieve local and global level intra-modal interaction; (iii) a Cross-modal Matching Network module, which applies bidirectional strategies, Textual-to-Visual and Visual-to-Textual matching, to implement bidirectional cross-modal interaction. Extensive experiments conducted on WikiMEL, RichpediaMEL, and WikiDiverse datasets demonstrate the outstanding performance of M3EL when compared to the state-of-the-art baselines.

cs.CV↗

arXiv2024

New insight into quantifying vacancy distribution in self-ion irradiated tungsten: a combined experimental and computational study

In this work, we propose a new approach based on positron annihilation spectroscopy to estimate the concentration of vacancy-type defects induced by self-ion irradiation in tungsten at room temperature, 500, and 700{\deg}C. Using experimental and Two-component density functional theory calculated annihilation characteristics of various vacancy clusters V$_{n}$ ($n$=1-65) and a positron trapping model associated with the simulated annealing algorithm, vacancy cluster concentration distribution could be extracted from experimental data. The method was validated against simulation results for room-temperature irradiation and transmission electron microscopy observations for higher temperatures. After irradiation at 500 and 700{\deg}C, small clusters (<20 vacancies, ~0.85 nm) undetectable by TEM were unveiled, with concentrations exceeding 10$^{25}$ m$^{-3}$, significantly higher than the concentration of TEM-visible defects (10$^{24}$ m$^{-3}$). Moreover, incorporating an oxygen-vacancy complex is deemed necessary to accurately replicate experimental data in samples subjected to high-temperature irradiation.

cond-mat.mtrl-sci↗

arXiv2024

High-dose long-time defect evolution in tungsten studied by atomistically informed Object Kinetic Monte Carlo simulations

Irradiation of materials in nuclear test reactors and power plants is known to alter the properties of the material. The irradiation event happening at pico- or nanosecond time scales are affecting the evolution and properties of the material on macroscopic timescales. Classical Molecular Dynamics simulations, which can capture the cascade event, are typically limited to nanosecond time scales, resulting in high dose rates. To achieve experimental dose rates, larger-scale models like Object Kinetic Monte Carlo are used, while they lack atomistic detail. The exact evolution of cascades in the vicinity of pre-existing defects is known to affect the defects formed, and the structure and morphology of the defects produced are crucial to know for determining macroscopic material behavior. Here we introduce a novel approach to integrate full Molecular Dynamics-based cascades into Object Kinetic Monte Carlo to achieve accurate dose rates, with the atomistic level accuracy of cascade overlap in tungsten. Our study reveals that incorporating full cascades significantly influences defect concentration levels. Not only is the concentration affected, but also the cluster statistics. We observe both that the full cascade can promote vacancy clustering at low temperatures and it can split existing voids at higher temperatures. These effects are missing in conventional Object Kinetic Monte Carlo simulations. This can be especially important in more complex materials, where many cascade-overlap effects are present.

cond-mat.mtrl-sci↗

arXiv2024

COTET: Cross-view Optimal Transport for Knowledge Graph Entity Typing

Knowledge graph entity typing (KGET) aims to infer missing entity type instances in knowledge graphs. Previous research has predominantly centered around leveraging contextual information associated with entities, which provides valuable clues for inference. However, they have long ignored the dual nature of information inherent in entities, encompassing both high-level coarse-grained cluster knowledge and fine-grained type knowledge. This paper introduces Cross-view Optimal Transport for knowledge graph Entity Typing (COTET), a method that effectively incorporates the information on how types are clustered into the representation of entities and types. COTET comprises three modules: i) Multi-view Generation and Encoder, which captures structured knowledge at different levels of granularity through entity-type, entity-cluster, and type-cluster-type perspectives; ii) Cross-view Optimal Transport, transporting view-specific embeddings to a unified space by minimizing the Wasserstein distance from a distributional alignment perspective; iii) Pooling-based Entity Typing Prediction, employing a mixture pooling mechanism to aggregate prediction scores from diverse neighbors of an entity. Additionally, we introduce a distribution-based loss function to mitigate the occurrence of false negatives during training. Extensive experiments demonstrate the effectiveness of COTET when compared to existing baselines.

cs.AI↗

arXiv2024

Leveraging Intra-modal and Inter-modal Interaction for Multi-Modal Entity Alignment

Multi-modal entity alignment (MMEA) aims to identify equivalent entity pairs across different multi-modal knowledge graphs (MMKGs). Existing approaches focus on how to better encode and aggregate information from different modalities. However, it is not trivial to leverage multi-modal knowledge in entity alignment due to the modal heterogeneity. In this paper, we propose a Multi-Grained Interaction framework for Multi-Modal Entity Alignment (MIMEA), which effectively realizes multi-granular interaction within the same modality or between different modalities. MIMEA is composed of four modules: i) a Multi-modal Knowledge Embedding module, which extracts modality-specific representations with multiple individual encoders; ii) a Probability-guided Modal Fusion module, which employs a probability guided approach to integrate uni-modal representations into joint-modal embeddings, while considering the interaction between uni-modal representations; iii) an Optimal Transport Modal Alignment module, which introduces an optimal transport mechanism to encourage the interaction between uni-modal and joint-modal embeddings; iv) a Modal-adaptive Contrastive Learning module, which distinguishes the embeddings of equivalent entities from those of non-equivalent ones, for each modality. Extensive experiments conducted on two real-world datasets demonstrate the strong performance of MIMEA compared to the SoTA. Datasets and code have been submitted as supplementary materials.

cs.IR↗

arXiv2024

HyperMono: A Monotonicity-aware Approach to Hyper-Relational Knowledge Representation

In a hyper-relational knowledge graph (HKG), each fact is composed of a main triple associated with attribute-value qualifiers, which express additional factual knowledge. The hyper-relational knowledge graph completion (HKGC) task aims at inferring plausible missing links in a HKG. Most existing approaches to HKGC focus on enhancing the communication between qualifier pairs and main triples, while overlooking two important properties that emerge from the monotonicity of the hyper-relational graphs representation regime. Stage Reasoning allows for a two-step reasoning process, facilitating the integration of coarse-grained inference results derived solely from main triples and fine-grained inference results obtained from hyper-relational facts with qualifiers. In the initial stage, coarse-grained results provide an upper bound for correct predictions, which are subsequently refined in the fine-grained step. More generally, Qualifier Monotonicity implies that by attaching more qualifier pairs to a main triple, we may only narrow down the answer set, but never enlarge it. This paper proposes the HyperMono model for hyper-relational knowledge graph completion, which realizes stage reasoning and qualifier monotonicity. To implement qualifier monotonicity HyperMono resorts to cone embeddings. Experiments on three real-world datasets with three different scenario conditions demonstrate the strong performance of HyperMono when compared to the SoTA.

cs.AI↗

arXiv2024

Continuous field tracking with machine learning and steady state spin squeezing

Entanglement plays a crucial role in proposals for quantum metrology, yet demonstrating quantum enhancement in sensing with sustained spin entanglement remains a challenging endeavor. Here, we combine optical pumping and continuous quantum nondemolition measurements to achieve a sustained spin squeezed state with $\bm{4 \times 10^{10}}$ hot atoms. A metrologically relevant steady state squeezing of $\bm{-3.23 \pm 0.24}$ dB using prediction and retrodiction is maintained for about one day. We employ the system to track different types of continuous time-fluctuating magnetic fields, where we construct deep learning models to decode the measurement records from the optical signals. Quantum enhancement due to the steady spin squeezing is verified in our atomic magnetometer. These results represent important progress towards applying long-lived quantum entanglement resources in realistic settings.

quant-ph↗