arXiv subjects

Shuang Wang

Publications and source records attributed to Shuang Wang.

At least 37 records · Page 2Linked to original sources

arXiv2026

Human-AI Co-Embodied Intelligence for Scientific Experimentation and Manufacturing

Scientific experimentation and manufacturing rely on prolonged protocol development and complex, multi-step implementation, which require continuous human expertise for precise execution and decision-making, limiting interpretability and scalability. Here, we introduce human-artificial intelligence (AI) co-embodied intelligence, a new form of physical AI that unites human researchers, agentic AI, and wearable hardware. In this paradigm, humans provide precise execution, while agentic AI contributes contextual reasoning, adaptive planning, and analysis. The wearable interface continuously captures experimentation and manufacturing, facilitating seamless communication between humans and AI. We instantiate this paradigm in a microfabrication cleanroom, leading to the agentic-physical experimentation (APEX) system which understands fabrication procedure with accuracy 51% higher than state-of-the-art multimodal large language models/vision language models (LLMs/VLMs), detects and corrects fabrication errors in real-time, and transfers procedural expertise to novice users. Critically, APEX system enables the co-development of fabrication protocols in cleanrooms, overcoming the incompatibility of elastomeric materials in standard microfabrication processes and enabling previously unattainable fabrication outcomes, as demonstrated by the wafer-scale realization of brain-level soft neural probe capable of single-unit-resolution neural recording. These results establish the human-AI co-embodied intelligence that extends agentic reasoning beyond computation into the physical domain, transforming scientific experimentation and manufacturing into autonomous, traceable, interpretable and scalable processes.

cs.AI↗

arXiv2026

Relativistic Position Verification with Coherent States

Determining the position of an entity is a fundamental prerequisite for nearly all activities. Classical means, however, have been proven incapable of providing secure position verification, meaning that a prover can mislead verifiers about its actual position. In this work, we propose and experimentally realize a secure position-verification protocol that leverages quantum optics and relativity within an information-theoretic framework. Using phase-randomized weak coherent states, two verifiers separated by 2 km securely verify the prover's position with an accuracy better than 75 meters. These results establish secure position-based authentication as a practical possibility, paving the way for applications in financial transactions, disaster response, and authenticated secure communications.

quant-ph↗

arXiv2025

5-GHz chip-based quantum key distribution with 1Mbps secure key rate over 150 km

Quantum key distribution (QKD) enables secure communication by harnessing the fundamental principles of quantum physics, which inherently guarantee information-theoretic security and intrinsic resistance to quantum computing attacks. However, the secure key rate of QKD typically decreases exponentially with increasing channel distance. In this work, by developing a novel polarization-state preparation method, an ultra-low time-jitter laser source and superconducting nanowire single-photon detectors, we demonstrate a 5-GHz integrated QKD system featuring ultra-low quantum bit error rates (QBERs). The system achieves secure key rates of 1.076 Mbps at 150 km and 105 kbps at 200 km over standard single-mode fiber channels, respectively. Our system substantially enhances the secure key rate, enabling high-resolution video calls with one-time-pad encryption over intercity backbone QKD links. This work represents a significant step forward in the development of high-performance practical QKD systems.

quant-ph↗

arXiv2025

Microcomb-driven large-scale fully connected quantum network

Fully connected quantum networks enable simultaneously connecting every user to every other user and are the most versatile and robust networking architecture. However, the scalability of such networks remains great challenge for practical applications. Here we construct a large-scale fully connected quantum network founded on two-photon Hong-Ou-Mandel (HOM) interference, where user-to-user security is guaranteed even with untrusted network provider. Using integrated soliton microcomb (SMC) and photonic encoding chips, we realize precise massive parallel frequency generation and locking, high-visibility HOM interferences and measurement-device-independent (MDI) quantum key distribution. The proposed architecture enables a 200-user fully connected quantum network over 200 kilometers with strict information-theoretic security via untrusted network provider. The implemented networking architecture paves the way for realizing large-scale fully connected MDI quantum networks across metropolitan and intercity regions.

quant-ph↗

arXiv2025

CLNet: Cross-View Correspondence Makes a Stronger Geo-Localizationer

Image retrieval-based cross-view geo-localization (IRCVGL) aims to match images captured from significantly different viewpoints, such as satellite and street-level images. Existing methods predominantly rely on learning robust global representations or implicit feature alignment, which often fail to model explicit spatial correspondences crucial for accurate localization. In this work, we propose a novel correspondence-aware feature refinement framework, termed CLNet, that explicitly bridges the semantic and geometric gaps between different views. CLNet decomposes the view alignment process into three learnable and complementary modules: a Neural Correspondence Map (NCM) that spatially aligns cross-view features via latent correspondence fields; a Nonlinear Embedding Converter (NEC) that remaps features across perspectives using an MLP-based transformation; and a Global Feature Recalibration (GFR) module that reweights informative feature channels guided by learned spatial cues. The proposed CLNet can jointly capture both high-level semantics and fine-grained alignments. Extensive experiments on four public benchmarks, CVUSA, CVACT, VIGOR, and University-1652, demonstrate that our proposed CLNet achieves state-of-the-art performance while offering better interpretability and generalizability.

cs.CV↗

arXiv2025

Reconstructing dark energy with model independent methods after DESI DR2 BAO

In this paper, we employ two model-independent approaches, including redshift binning method and polynomial interpolation method, to reconstruct dark energy (DE) equation of state (EoS) $w(z)$ and DE density function $f(z)$. Our analysis incorporates data from the Dark Energy Spectroscopic Instrument (DESI) data release 2, Cosmic Microwave Background (CMB) distance priors from Planck 2018 and Atacama Cosmology Telescope data release 6, and three Type Ia supernovae (SN) compilations (PantheonPlus, Union3, and DESY5). To ensure model independence, we adopt three redshift binning schemes (n=3, 4, 5) and three polynomial interpolation schemes with the same number of nodes (n=3, 4, 5). Our main conclusions are as follows: 1) After taking into account DESI data, there is a trend that DE should evolve with redshift (with deviations from the cosmological constant reaching at least a $2.13σ$ confidence level), indicating that current observations favor a dynamical DE. 2) In the redshift range $0.5 < z < 1.5$, the DE EoS w(z) exhibits a decreasing trend and crosses the phantom divide $w=-1$, suggesting quintom-like behavior. 3) The DE density f(z) first increases at low redshift, reaching a hump around $z\approx 0.5$, and then decreases at $0.5 < z < 1.5$, with a rapid decrease at $z>1.0$. 4) For $z > 1.5$, current data are insufficient to place strong constraints on the evolution of DE, resulting in large uncertainties in the DE reconstruction. It must be emphasized that, these four main conclusions are independent of specific reconstruction models, and are insensitive to the choice of SN compilations.

astro-ph.CO↗

arXiv2025

ReFineG: Synergizing Small Supervised Models and LLMs for Low-Resource Grounded Multimodal NER

Grounded Multimodal Named Entity Recognition (GMNER) extends traditional NER by jointly detecting textual mentions and grounding them to visual regions. While existing supervised methods achieve strong performance, they rely on costly multimodal annotations and often underperform in low-resource domains. Multimodal Large Language Models (MLLMs) show strong generalization but suffer from Domain Knowledge Conflict, producing redundant or incorrect mentions for domain-specific entities. To address these challenges, we propose ReFineG, a three-stage collaborative framework that integrates small supervised models with frozen MLLMs for low-resource GMNER. In the Training Stage, a domain-aware NER data synthesis strategy transfers LLM knowledge to small models with supervised training while avoiding domain knowledge conflicts. In the Refinement Stage, an uncertainty-based mechanism retains confident predictions from supervised models and delegates uncertain ones to the MLLM. In the Grounding Stage, a multimodal context selection algorithm enhances visual grounding through analogical reasoning. In the CCKS2025 GMNER Shared Task, ReFineG ranked second with an F1 score of 0.6461 on the online leaderboard, demonstrating its effectiveness with limited annotations.

cs.IR↗

arXiv2025

Fused Partial Gromov-Wasserstein for Structured Objects

Structured data, such as graphs, is vital in machine learning due to its capacity to capture complex relationships and interactions. In recent years, the Fused Gromov-Wasserstein (FGW) distance has attracted growing interest because it enables the comparison of structured data by jointly accounting for feature similarity and geometric structure. However, as a variant of optimal transport (OT), classical FGW assumes an equal mass constraint on the compared data. In this work, we relax this mass constraint and propose the Fused Partial Gromov-Wasserstein (FPGW) framework, which extends FGW to accommodate unbalanced data. Theoretically, we establish the relationship between FPGW and FGW and prove the metric properties of FPGW. Numerically, we introduce Frank-Wolfe solvers and Sinkhorn solvers for the proposed FPGW framework. Finally, we evaluate the FPGW distance through graph matching, graph classification and graph clustering experiments, demonstrating its robust performance.

cs.LG↗

arXiv2025

Fast qubit-based frequency recovery algorithm for quantum key distribution

Clock synchronization serves as a foundational subsystem in quantum key distribution (QKD). The recently proposed Qubit-based synchronization (Qubit4Sync) has opportunities in eliminating additional cost, noise, and potential side channels. It offers a promising alternative to dedicated synchronization hardware. However, the current frequency recovery process in Qubit4Sync requires high data throughput and computational speed, limiting practical use. To overcome these issues, we developed a fast frequency recovery algorithm that increases the recovery rate by orders of magnitude and remains robust under bad signal-to-noise ratio (SNR). This enables Qubit4Sync to operate effectively in mainstream gated-mode QKD systems. We further establish a theoretical model for frequency recovery, showing that our algorithm is robust against disturbances like dead time, jitter, and afterpulse. A frequency-domain SNR calculation method is also provided to guide parameter design for specific experimental conditions. This work opens the door to practical Qubit4Sync deployment in general QKD systems.

quant-ph↗

arXiv2025

A hybrid-frequency on-chip programmable synthetic-dimension simulator with arbitrary couplings

High-performance photonic chips provide a powerful platform for analog computing, enabling the simulation of high-dimensional physical systems using low-dimensional devices with additional synthetic dimensions. The realization of large-scale complex simulations necessitates an architecture capable of arbitrary coupling configurations (encompassing symmetric, asymmetric and long-range coupling schemes) which is also crucial for scaling up. Previous approaches rely on excessive physical components to introduce asymmetric coupling, however, are restricted in reconfiguring and scaling by the relatively complicated structures. Here, to solve this problem, we propose a hybrid-frequency synthetic-dimension simulator architecture that combines both intra-resonant and inter-resonant frequency-lattice sites, and experimentally demonstrate it using the thin-film lithium niobate (TFLN) photonic chip. Employing this hybrid programmable architecture, we are able to simulate both the regular and long-range coupled forms of diverse compound-lattice models, such as the Hall ladder, Creutz ladder (symmetric) and Su-Schrieffer-Heeger (SSH, asymmetric) model, on a single chip, simultaneously reducing the experimental requirements significantly. As results, the direct readout of the bandstructure of the SSH model is able to be achieved, to be distinguished from all previous works, and important phenomena such as spin-momentum locking, topological flat band and Aharonov-Bohm cage effect are also observed with lower experimental requirements. Furthermore, applications like piecewise-continuous optical frequency shifting can be enabled by cascading our devices. Our results offer promising insights for future large-scale complex on-chip simulators with arbitrary couplings.

physics.optics↗

arXiv2025

Versatile photonic frequency synthetic dimensions using a single Mach-Zehnder-interferometer-assisted device on thin-film lithium niobate

Investigating physical models with photonic synthetic dimensions has been generating great interest in vast fields of science. The rapid developing thin-film lithium niobate (TFLN) platform, for its numerous advantages including high electro-optic coefficient and scalability, is well compatible with the realization of synthetic dimensions in the frequency together with spatial domain. While coupling resonators with fixed beam splitters is a common experimental approach, it often lacks tunability and limits coupling between adjacent lattices to sites occupying the same frequency domain positions. Here, on the contrary, we conceive the resonator arrays connected by electro-optic tunable Mach-Zehnder interferometers in our configuration instead of fixed beam splitters. By applying bias voltage and RF modulation on the interferometers, our design extends such coupling to long-range scenario and allows for continuous tuning on each coupling strength and synthetic effective magnetic flux. Therefore, our design enriches controllable coupling types that are essential for building programmable lattice networks and significantly increases versatility. As the example, we experimentally fabricate a two-resonator prototype on the TFLN platform, and on this single chip we realize well-known models including tight-binding lattices, topological Hall ladder and Creutz ladder. We directly observe the band structures in the quasi-momentum space and important phenomena such as spin-momentum locking and the Aharonov-Bohm cage effect. These results demonstrate the potential for convenient simulations of more complex models in our configuration.

physics.optics↗

arXiv2025

Experimental demonstration of drone-based quantum key distribution

Quantum state transferring has been demonstrated using drones via entanglement distribution. Here we demonstrate the first drone-based quantum task, for quantum key distribution (QKD). Compact and polarization-maintaining acquisition, pointing, and tracking system and QKD modules are developed and loaded on a home-made octocopter, within takeoff weight of 30 kg. Real-time QKD is performed over 200 m distance with 8.48 kHz average secret key rate, using polarization-coded decoy-state BB84 protocol. With the capability of secret key distribution using a drone, wireless communication can be expected with enhanced security in the quantum approach, between mobile nodes towards a network.

quant-ph↗

arXiv2025

Domain-aware Category-level Geometry Learning Segmentation for 3D Point Clouds

Domain generalization in 3D segmentation is a critical challenge in deploying models to unseen environments. Current methods mitigate the domain shift by augmenting the data distribution of point clouds. However, the model learns global geometric patterns in point clouds while ignoring the category-level distribution and alignment. In this paper, a category-level geometry learning framework is proposed to explore the domain-invariant geometric features for domain generalized 3D semantic segmentation. Specifically, Category-level Geometry Embedding (CGE) is proposed to perceive the fine-grained geometric properties of point cloud features, which constructs the geometric properties of each class and couples geometric embedding to semantic learning. Secondly, Geometric Consistent Learning (GCL) is proposed to simulate the latent 3D distribution and align the category-level geometric embeddings, allowing the model to focus on the geometric invariant information to improve generalization. Experimental results verify the effectiveness of the proposed method, which has very competitive segmentation accuracy compared with the state-of-the-art domain generalized point cloud methods.

cs.CV↗

arXiv2025

Measurement-device-independent quantum key distribution with asymmetric sources

Measurement-device-independent quantum key distribution (MDI-QKD), which eliminates all the attacks from the eavesdropper to the measurement party, has been one of the most promising technology for the implementation of end-to-end quantum networks. In practice, the asymmetry of both sources and channels is generally inevitable. Therefore, we propose a theory to analyze the performance when any two MDI users in networks communicates using asymmetric sources in distinct single or multiple temporal modes. As a specific application, we model to obtain the key rate of MDI-QKD with weak coherent pulse source and spontaneous parametric down-conversion source, and compare the performance to the cases with symmetric (i.e. identical) sources. The result demonstrates that the actual performance does not degrade due to the asymmetry of sources. In contrary, it maintains at a good level over the entire distance we study. This work provides a theoretical basis for analyzing and optimizing MDI-QKD networks with asymmetric sources, and thus paving the way for the practical deployment of completely asymmetric MDI-QKD networks.

quant-ph↗

arXiv2025

EEG-based Epileptic Prediction via a Two-stage Channel-aware Set Transformer Network

Epilepsy is a chronic, noncommunicable brain disorder, and sudden seizure onsets can significantly impact patients' quality of life and health. However, wearable seizure-predicting devices are still limited, partly due to the bulky size of EEG-collecting devices. To relieve the problem, we proposed a novel two-stage channel-aware Set Transformer Network that could perform seizure prediction with fewer EEG channel sensors. We also tested a seizure-independent division method which could prevent the adjacency of training and test data. Experiments were performed on the CHB-MIT dataset which includes 22 patients with 88 merged seizures. The mean sensitivity before channel selection was 76.4% with a false predicting rate (FPR) of 0.09/hour. After channel selection, dominant channels emerged in 20 out of 22 patients; the average number of channels was reduced to 2.8 from 18; and the mean sensitivity rose to 80.1% with an FPR of 0.11/hour. Furthermore, experimental results on the seizure-independent division supported our assertion that a more rigorous seizure-independent division should be used for patients with abundant EEG recordings.

eess.SP↗

arXiv2025

Secure quantum key distribution against correlated leakage source

Quantum key distribution (QKD) provides information theoretic security based on quantum mechanics, however, its practical deployment is challenged by imperfections of source devices. Among various source loopholes, correlations between transmitted pulses pose a significant yet underexplored security risk, potentially compromising QKD's theoretical guarantees. In this work, we propose a security analysis framework for QKD under correlations, enabling finite-key analysis for the first time by extending and rearranging QKD rounds and leveraging the generalized chain rule. Based on this framework, and inspired by the idea of side-channel-secure QKD, we develop a secure QKD against correlated leakage source only need the characterization of correlation range and the lower bound on the vacuum component of the prepared states. Additionally, our framework can be extended to other QKD protocols, offering a general approach to consider correlation induced security vulnerabilities. The simulation results demonstrate the effectiveness of our protocol and its significantly superior tolerance to imperfect parameters compared to existing protocols. This work provides a crucial step toward closing security loopholes in QKD, enhancing its practicality, and ensuring long-distance,high-performance secure communication under real-world constraints.

quant-ph↗

arXiv2025

City-Level Foreign Direct Investment Prediction with Tabular Learning on Judicial Data

To advance the United Nations Sustainable Development Goal on promoting sustained, inclusive, and sustainable economic growth, foreign direct investment (FDI) plays a crucial role in catalyzing economic expansion and fostering innovation. Precise city-level FDI prediction is quite important for local government and is commonly studied based on economic data (e.g., GDP). However, such economic data could be prone to manipulation, making predictions less reliable. To address this issue, we try to leverage large-scale judicial data which reflects judicial performance influencing local investment security and returns, for city-level FDI prediction. Based on this, we first build an index system for the evaluation of judicial performance over twelve million publicly available adjudication documents according to which a tabular dataset is reformulated. We then propose a new Tabular Learning method on Judicial Data (TLJD) for city-level FDI prediction. TLJD integrates row data and column data in our built tabular dataset for judicial performance indicator encoding, and utilizes a mixture of experts model to adjust the weights of different indicators considering regional variations. To validate the effectiveness of TLJD, we design cross-city and cross-time tasks for city-level FDI predictions. Extensive experiments on both tasks demonstrate the superiority of TLJD (reach to at least 0.92 R2) over the other ten state-of-the-art baselines in different evaluation metrics.

cs.AI↗

arXiv2025

A comprehensive numerical study on four categories of holographic dark energy models

Holographic dark energy (HDE), which arises from a theoretical attempt to apply the holographic principle (HP) to the dark energy (DE) problem, has attracted significant attention over the past two decades. We perform a comprehensive numerical study on HDE models that can be classified into four categories: 1) HDE models with other characteristic length scale, 2) HDE models with extended Hubble scale, 3) HDE models with dark sector interaction, 4) HDE models with modified black hole entropy. For theoretical models, we select seven representative models, including the original HDE (OHDE) model, Ricci HDE (RDE) model, generalized Ricci HDE (GRDE) model, interacting HDE (IHDE1 and IHDE2) models, Tsallis HDE (THDE) model, and Barrow HDE (BHDE) model. For cosmological data, we use the Baryon Acoustic Oscillation (BAO) data from the Dark Energy Spectroscopic Instrument (DESI) 2024 measurements, the Cosmic Microwave Background (CMB) distance priors data from the Planck 2018, and the type Ia supernovae (SNe) data from the PantheonPlus compilation. Using $χ^2$ statistic and Bayesian evidence, we compare these HDE models with current observational data. It is found that: 1) The $Λ$CDM remains the most competitive model, while the RDE model is ruled out. 2) HDE models with dark sector interaction perform the worst across the four categories, indicating that the interaction term is not favored under the framework of HDE. 3) The other three categories show comparable performance. The OHDE model performs better in the BAO+CMB dataset, and the HDE models with modified black hole entropy perform better in the BAO+CMB+SN dataset. 4) HDE models with the future event horizon exhibit significant discrepancies in parameter space across datasets. The BAO+CMB dataset favors a phantom-like HDE, whereas the BAO+CMB+SN leads to an equation of state (EoS) much closer to the cosmological constant.

astro-ph.CO↗