Searcharxiv⌕ Search

arXiv subjects

Mingyuan Wang

Publications and source records attributed to Mingyuan Wang.

At least 19 recordsLinked to original sources
arXiv2026

GHZ-Preserving Gates and Optimized Distillation Circuits

Greenberger-Horne-Zeilinger (GHZ) states play a central role in quantum computing and communication protocols, as a typical multipartite entanglement resource. This work introduces an efficient enumeration and simulation method for circuits that preserve and distill noisy GHZ states, significantly reducing the simulation complexity of a gate on $n$ qubits, from exponential $O(2^n)$ for standard state-vector methods or $O(n)$ for Clifford circuits, to a constant $O(1)$ for the method presented here. This method has profound implications for the design of quantum networks, where preservation and purification of entanglement with minimal resource overhead is critical. In particular, we demonstrate the use of the new method in an optimization procedure enabled by the fast simulation, that discovers GHZ distillation circuits far outperforming the state of the art. Fine-tuning to arbitrary noise models is possible as well. We also show that the method naturally extends to graph states that are local Clifford equivalent to GHZ states.

quant-ph↗
arXiv2026

First Observation of Fishbone-Driven Zonal Flows with Fine Reversed Structure in Tokamak Plasmas

We present the first direct experimental observation of fishbone-driven zonal flows in the core of the EAST tokamak. In contrast to the global pattern predicted by previous models and simulations based on the energetic-particle-expulsion mechanism, the observed flows exhibit a fine-scale, radially reversed structure inside the q = 1 rational surface. The flow rises faster and saturates earlier than the fishbone within a single burst, indicating that a beat-driven nonlinear process dominates the early stage rather than the energetic-particle-expulsion mechanism. Global nonlinear gyrokinetic simulations quantitatively reproduce the observed radial profile and reveal that this structure arises from the cancellation of comparable but opposite contributions from thermal ions and electrons. This cancellation mechanism is not captured in previous theoretical frameworks. These findings establish that the fishbone can generate sheared flows with a distinct radial topology, offering a promising pathway for regulating turbulence and improving core confinement.

physics.plasm-ph↗
arXiv2026

A unified quenching model in Geant 4 simulation for $α$, proton and electron particles in liquid scintillator detectors

The liquid scintillator (LS) detectors are widely used in reactor neutrino experiments. To precisely measure the oscillated neutrino spectrum, it is crucial to understand the LS energy response, especially the quenching effect. Numerous bench-top measurements have been conducted on quenching effects for $α$ particles, protons, and electrons. These results have typically been described by Birks' law, but with different Birks' coefficients required for different particle species. In this study, we find that if more secondary electrons are allowed to be generated in Geant4 simulations, the bench-top results for all particles can be well fitted using a single Birks' coefficient of about $0.013~{\rm g\,cm^{-2}\,MeV^{-1}}$. The underlying reason is that a large fraction of the primary energy is deposited through the generation of $δ$-electrons with energies below 4~keV, which should be tracked separately due to their different quenching behavior compared to primary particles. This study provides a unified framework for comparing Birks' coefficients from different bench-top measurements and helps experiments like JUNO and SNO+ better tune their simulations for precision energy measurements.

hep-ex↗
arXiv2026

Relativistic Time Scales and Transformations in the Solar System

Each solar-system observable is characterised by celestial reference system (CRS) coordinate time, proper time on its world line, and the transformation between them. Ephemerides and Deep Space Network (DSN) tracking use the International Astronomical Union (IAU) barycentric and body-centric hierarchy, now extended to cislunar and Mars work. The IERS Conventions, Moyer radiometric models, and recent lunar-time papers distribute metric, scale, and tracking formulae across separate manuals. Merged Chang'e- or Tianwen-class data can acquire microsecond-level range and Doppler biases unless proper time $τ$ is mapped consistently to barycentric and body-centric coordinate times. We present a unified 1PN documentation chain: tabulated harmonic Christoffel symbols through $\mathcal{O}(c^{-4})$, the barycentric-geocentric-terrestrial coordinate-time sequence, Fermi normal coordinates, null-geodesic observables, and a 1PN two-way range-rate expansion, applied in parallel to Mars (MCRS/MCG) and lunar (LCRS/TCL) body-centric systems. The chain yields a Mars areoid-geoid metric clock-rate difference of $\sim$48~$μ$s\,day$^{-1}$ and lunar selenoid-geoid rates of $\sim$57.4-58.7~$μ$s\,day$^{-1}$ consistent with published nested coefficients. Mars-range Shapiro-rate terms reach $10^{-12}$-$10^{-13}$. Multi-CRS consistency relies on documented transformation chains rather than a single master clock.

astro-ph.EP↗
arXiv2026

MSFA-Net: An Advanced Deep Learning Model for Identifying Blue Horizontal-Branch Stars from LAMOST DR12

Blue horizontal-branch (BHB) stars are low-mass, core helium-burning objects with nearly constant luminosities, making them powerful tracers of old, metal-poor populations and valuable standard candles for mapping the Galactic halo. However, robustly identifying BHB stars from low-resolution spectra remains challenging. We present MSFA-Net, a two-stage framework developed for the Large Sky Area Multi-Object Fiber Spectroscopic Telescope (LAMOST) DR12. By combining multi-scale convolutions with a soft frequency attention mechanism, MSFA-Net learns discriminative representations in both the wavelength domain and the Fourier-frequency domain. On the test set, the framework achieves a precision of 94.67% in the initial multiclass screening and 98.07% in the subsequent binary refinement. Applying the pipeline to LAMOST DR12, we retrieve 27,853 BHB candidate spectra. After spectral deduplication and removal of previously known objects, we identify 3583 new BHB stars, confirmed via Balmer-line profile fitting. We further estimate atmospheric parameters (Teff, log g, and [Fe/H]) using the machine-learning-based SLAM model and examine their distributions. A non-negligible subset shows unusually high log g and/or metallicities, which we interpret primarily as inference-related systematics rather than intrinsic properties. Photometric cross-matching with Gaia DR3 and color-magnitude diagrams provide an additional consistency check for the sample. The resulting catalog substantially enlarges the spectroscopically confirmed BHB sample from LAMOST and offers a homogeneous data set for studies of Galactic-halo structure and stellar populations.

astro-ph.SR↗
arXiv2026

TuringViT: Making SOTA Vision Transformers Accessible to All

Modern VLMs and VLA systems commonly adopt off-the-shelf ViTs such as SigLIP2 as visual encoders, but diverse downstream requirements in latency, temporal modeling, and VLM integration often call for customized SOTA-level ViTs. Training such encoders remains beyond the reach of much of the community, as it requires massive image-text data, while standard softmax attention makes high-resolution or dynamic-resolution pretraining prohibitively costly and often forces low-resolution pretraining followed by post-hoc adaptation. TuringViT addresses these challenges with three key designs: Turing Linear Attention (TLA) for efficient sequence modeling, VISTA-Curation to construct supervision-rich image-video training data, and native dynamic-resolution pretraining that supports flexible inputs from the start and transfers seamlessly to downstream VLMs. As a result, TuringViT outperforms leading open-source ViT baselines with only 10% of the data, achieves stronger downstream VLM performance, and delivers substantially better latency scaling on high-resolution inputs. Our scaling-law analysis further shows that TuringViT continues to improve predictably with curated data scale, far from saturation. Its fast adaptation, hardware-friendly design, and efficient deployment have made it a unified visual foundation across XPeng's AI systems. More broadly, TuringViT provides a reproducible pipeline that dramatically lowers the cost for the community to train, customize, and deploy SOTA-level ViTs, moving toward making such Vision Transformers accessible to all.

cs.CV↗
arXiv2026

Efficient Graph State Purification with Factorized Graph-Preserving Operations across Local Clifford Orbits

Graph states form a broad class of multipartite entangled states underlying measurement-based quantum computation, quantum networks, and stabilizer codes. However, systematic entanglement distillation for arbitrary graph states remains challenging because the circuit design space grows rapidly with the number of parties. We introduce a group of Clifford operations that we call "factorized graph-preserving". It enables us to efficiently enumerate and optimize graph-state purification circuits at finite size for realistic noisy hardware. These operations map products of graph-basis states to products of graph-basis states, so their action can be represented as permutations of graph-basis labels. Moreover, this useful gate set admits a compact factorized description determined by simple graph-theoretic features. This structure also allows, after some initial cached precomputation, drastically lower computational complexity for simulating a gate. We further organize these operations over local-complementation (LC) orbits using minimum-edge representatives (MERs), which let us design purification circuits that apply to all locally equivalent graph states (up to a basis change). Using this framework, we optimize noisy finite-size multipartite distillation circuits for several graph-state families. Numerical results show that the resulting graph-preserving circuits can outperform standard recurrence-based purification protocols under realistic gate and measurement noise. Our results establish LC-orbit structure and factorized graph-preserving operations as practical tools for scalable, topology-aware and hardware-constrained graph-state distillation protocol design. Our work can also be interpreted as a graph-based heuristic for finding transversal gates.

quant-ph↗
arXiv2026

Bypassing Prompt Guards in Production with Controlled-Release Prompting

Ball et al. recently established that prompt filtering for AI alignment faces a fundamental barrier: under standard cryptographic assumptions, no filter running significantly faster than the protected model can universally distinguish adversarial prompts from benign ones. We investigate whether this impossibility result translates to real-world vulnerabilities in deployed large language model (LLM) systems. We answer affirmatively by introducing controlled-release prompting, a practical instantiation of the theoretical framework that exploits the resource asymmetry between lightweight input filters and the main models they protect. Unlike the theoretical construction, our attack does not require model modification: it generates malicious prompts that are indecipherable by any bounded filter yet remain tractable to the target LLM. We find our attack to be successful on four major chat platforms (Google Gemini, DeepSeek Chat, xAI Grok, and Mistral Le Chat) where baseline methods fail. Additionally, we apply our attack to extract copyrighted data from Gemini. Finally, we provide a systematic evaluation of 14 open-weight prompt guard models, revealing that even reasoning-capable filters cannot reliably detect our attack without incurring prohibitive resource overhead.

cs.LG↗
arXiv2026

Exploring Vision Neural Network Pruning via Screening Methodology

The remarkable performance of modern deep neural networks (DNNs) is largely driven by their massive scale, often comprising tens to hundreds of millions-or even billions-of parameters. However, such a scale incurs substantial storage and computational costs, hindering deployment on platforms such as edge devices that require energy-efficient and real-time processing. In this paper, we propose a network pruning framework that reduces both storage and computation requirements by an order of magnitude while preserving model accuracy. Our approach eliminates non-essential parameters through a statistical analysis of component significance across classification categories. Specifically, we employ a F-statistic-based screening technique combined with a weighted evaluation scheme to quantify the contributions of connections and channels, enabling both unstructured and structured pruning within a unified framework. Extensive experiments on real-world vision datasets, covering both fully connected neural networks (FNNs) and convolutional neural networks (CNNs), demonstrate that the proposed framework produces compact and efficient models that are highly competitive with the state of art apporoaches.

cs.LG↗
arXiv2026

FineServe: A Fine-Grained Dataset and Characterization of Global LLM Serving Workloads

Large language models (LLMs) are increasingly deployed as always-on online services, making efficient LLM serving a critical systems challenge. Achieving low latency and high throughput under volatile demand requires deep understanding of real-world serving workloads, yet existing studies often rely on proxy traces or coarse-grained characterizations that fail to capture the heterogeneity of modern multi-model LLM platforms. We present FineServe, an in-the-wild, multi-model LLM serving workload dataset collected from a global commercial marketplace, enabling fine-grained characterization of real-world serving dynamics across heterogeneous models and tasks. Leveraging FineServe, we conduct a comprehensive analysis of arrival dynamics and token behavior, revealing fundamentally different fluctuation regimes across model architectures, scales and task intents. Building on these insights, we develop the FineServe workload generator, which composes fine-grained model-aware workloads into configurable mixtures tailored for benchmarking multi-model serving platforms. By exposing these fine-grained workload dynamics, FineServe provides a realistic foundation for evaluating routing, scheduling, and capacity-planning strategies in LLM serving systems. FineServe is available at https://github.com/hihiztc1/FineServe.

cs.AI↗
arXiv2026

Joint Representation Learning and Clustering via Gradient-Based Manifold Optimization

Clustering and dimensionality reduction have been crucial topics in machine learning and computer vision. Clustering high-dimensional data has been challenging for a long time due to the curse of dimensionality. For that reason, a more promising direction is the joint learning of dimension reduction and clustering. In this work, we propose a Manifold Learning Framework that learns dimensionality reduction and clustering simultaneously. The proposed framework is able to jointly learn the parameters of a dimension reduction technique (e.g. linear projection or a neural network) and cluster the data based on the resulting features (e.g. under a Gaussian Mixture Model framework). The framework searches for the dimension reduction parameters and the optimal clusters by traversing a manifold,using Gradient Manifold Optimization. The obtained The proposed framework is exemplified with a Gaussian Mixture Model as one simple but efficient example, in a process that is somehow similar to unsupervised Linear Discriminant Analysis (LDA). We apply the proposed method to the unsupervised training of simulated data as well as a benchmark image dataset (i.e. MNIST). The experimental results indicate that our algorithm has better performance than popular clustering algorithms from the literature.

stat.ML↗
arXiv2026

Impact of the $^5$Li resonance in $α$-$p$ elastic scattering on precision measurements of neutrino oscillation parameters

Precision measurements of four neutrino oscillation parameters, $θ_{12}$, $θ_{13}$, $Δm^2_{21}$, and |$Δm^2_{31}$|, face significant interference from a previously overlooked correlated background. Recent findings from the SNO+ and JUNO experiments reveal that cascade decays of $^{214}$Bi-$^{214}$Po in liquid scintillator detectors can mimic inverse beta decay signals from reactor and geoneutrinos, with a misidentification probability on the order of $10^{-4}$ when hydrogen neutron capture is used, a rate ten times higher than Geant4 simulations predicted. This work identifies the $^5$Li resonance in $α$-$p$ elastic scattering as the underlying cause. For alpha energies above 5~MeV, the cross section is hundreds of times larger than that of Rutherford scattering. After correctly incorporating the differential cross section into Geant4, the misidentification probability is recalculated as 1.9$\times$10$^{-4}$. The simulated shape of the long tail in the alpha deposited energy also differs from the extrapolation models currently used by SNO+ and JUNO. These results will assist both experiments in more accurately estimating this novel background, thereby refining measurements of neutrino oscillation parameters and the geoneutrino flux. Additionally, the study implies an overlooked background with a rate of 0.5 events per detector per day in the Daya Bay $θ_{13}$ analysis using hydrogen neutron capture, leading to an increase of $\sin^22θ_{13}$ by approximately 0.012. Consequently, the Particle Data Group's reported $\sin^2θ_{13}$ value shall increase by about 0.006~(1$σ$).

nucl-ex↗
arXiv2025

MCANet: A Coherent Multimodal Collaborative Attention Network for Advanced Modulation Recognition in Adverse Noisy Environments

As wireless communication systems evolve, automatic modulation recognition (AMR) plays a key role in improving spectrum efficiency, especially in cognitive radio systems. Traditional AMR methods face challenges in complex, noisy environments, particularly in low signal-to-noise ratio (SNR) conditions. This paper introduces MCANet (Multimodal Collaborative Attention Network), a multimodal deep learning framework designed to address these challenges. MCANet employs refined feature extraction and global modeling to support its fusion strategy.Experimental results across multiple benchmark datasets show that MCANet outperforms mainstream AMR models, offering better robustness in low-SNR conditions.

eess.SP↗
arXiv2025

Black-Box Crypto is Useless for Pseudorandom Codes

A pseudorandom code is a keyed error-correction scheme with the property that any polynomial number of encodings appear random to any computationally bounded adversary. We show that the pseudorandomness of any code tolerating a constant rate of random errors cannot be based on black-box reductions to almost any generic cryptographic primitive: for instance, anything that can be built from random oracles, generic multilinear groups, and virtual black-box obfuscation. Our result is optimal, as Ghentiyala and Guruswami (2024) observed that pseudorandom codes tolerating any sub-constant rate of random errors exist using a black-box reduction from one-way functions. The key technical ingredient in our proof is the hypercontractivity theorem for Boolean functions, which we use to prove our impossibility in the random oracle model. It turns out that this easily extends to an impossibility in the presence of ``crypto oracles,'' a notion recently introduced -- and shown to be capable of implementing all the primitives mentioned above -- by Lin, Mook, and Wichs (EUROCRYPT 2025).

cs.CR↗
arXiv2025

Temporal variability in low-frequency radio interference: Insight from high-cadence monitoring at a candidate radio notification zone in Malaysia

Extensive radio frequency interference (RFI) monitoring is essential in the site selection process before constructing radio astronomy observatories, followed by mitigation strategies to minimize its adverse effects. Malaysia has an enormous prospect for radio astronomy due to its prominent location in the centre of Southeast Asia, but is challenged by its relatively high population density. In this research article, we perform high-cadence, low-frequency RFI monitoring at two sites, each representing an urban and a rural environment. Using modified generalized spectral kurtosis (GSK) as an RFI detection method, we ascertain the suitability of Glami Lemi, a rural area in the centre of Peninsular Malaysia previously assigned as a candidate radio notification zone (RNZ), as a potential site for radio astronomy observations due to its lower RFI contamination in our high-cadence monitoring, especially when compared with urban areas. We identified a number of persistent and transient RFI in our dataset, associate each of them with their potential origins and, if present, characterize their temporal evolution. A few types of RFI mitigation strategies were also tested and discussed. This study lays the groundwork for Malaysia's endeavours in establishing its first research-grade radio telescope, emphasizing the importance of robust RFI detection and mitigation strategies in optimizing observational outcomes.

astro-ph.IM↗
arXiv2025

Technical Report of TeleChat2, TeleChat2.5 and T1

We introduce the latest series of TeleChat models: \textbf{TeleChat2}, \textbf{TeleChat2.5}, and \textbf{T1}, offering a significant upgrade over their predecessor, TeleChat. Despite minimal changes to the model architecture, the new series achieves substantial performance gains through enhanced training strategies in both pre-training and post-training stages. The series begins with \textbf{TeleChat2}, which undergoes pretraining on 10 trillion high-quality and diverse tokens. This is followed by Supervised Fine-Tuning (SFT) and Direct Preference Optimization (DPO) to further enhance its capabilities. \textbf{TeleChat2.5} and \textbf{T1} expand the pipeline by incorporating a continual pretraining phase with domain-specific datasets, combined with reinforcement learning (RL) to improve performance in code generation and mathematical reasoning tasks. The \textbf{T1} variant is designed for complex reasoning, supporting long Chain-of-Thought (CoT) reasoning and demonstrating substantial improvements in mathematics and coding. In contrast, \textbf{TeleChat2.5} prioritizes speed, delivering rapid inference. Both flagship models of \textbf{T1} and \textbf{TeleChat2.5} are dense Transformer-based architectures with 115B parameters, showcasing significant advancements in reasoning and general task performance compared to the original TeleChat. Notably, \textbf{T1-115B} outperform proprietary models such as OpenAI's o1-mini and GPT-4o. We publicly release \textbf{TeleChat2}, \textbf{TeleChat2.5} and \textbf{T1}, including post-trained versions with 35B and 115B parameters, to empower developers and researchers with state-of-the-art language models tailored for diverse applications.

cs.CL↗
arXiv2025

Utilizing Optic Fiber Interferometry in Forced Vibration Experimentation for Educational Purposes

This study introduces an experimental teaching method that employs optic fiber interferometry (OFI) to investigate forced vibration phenomena. It is designed for undergraduate physics majors with foundational mechanics and optics training and optics-focused graduate students. This approach aims to deepen students' understanding of forced vibration theory and interferometric measurement principles while fostering skills in experimental design, data analysis, and problem solving. Leveraging OFI's high-precision displacement measurement capabilities, the experiment enabled accurate tracking of frequency and displacement variations. By scanning the driving force frequency, students obtained amplitude frequency curves to determine the system's natural frequency, which closely aligned with theoretical predictions. This method may bridge theoretical concepts and practical applications, offering insights into teaching vibration theory and precision measurement techniques and equipping students with integrated knowledge for real-world challenges.

physics.ed-ph↗
arXiv2025

Overview of EXL-50 Research Progress and Future Plan

XuanLong-50 (EXL-50) is the first medium-size spherical torus (ST) in China, with the toroidal field at major radius at 50 cm around 0.5T. CS-free and non-inductive current drive via electron cyclotron resonance heating (ECRH) was the main physics research issue for EXL-50. Discharges with plasma currents of 50 kA - 180 kA were routinely obtained in EXL-50, with the current flattop sustained for up to or beyond 2 s. The current drive effectiveness on EXL-50 was as high as 1 A/W for low-density discharges using 28GHz ECRH alone for heating power less than 200 kW. The plasma current reached Ip>80 kA for high-density (5*10e18m-2) discharges with 150 kW 28GHz ECRH. Higher performance discharge (Ip of about 120 kA and core density of about 1*10e19m-3) was achieved with 150 kW 50GHz ECRH. The plasma current in EXL-50 was mainly carried by the energetic electrons.Multi-fluid equilibrium model has been successfully applied to reconstruct the magnetic flux surface and the measured plasma parameters of the EXL-50 equilibrium. The physics mechanisms for the solenoid-free ECRH current drive and the energetic electrons has also been investigated. Preliminary experimental results show that 100 kW of lower hybrid current drive (LHCD) waves can drive 20 kA of plasma current. Several boron injection systems were installed and tested in EXL-50, including B2H6 gas puffing, boron powder injection, boron pellet injection. The research plan of EXL-50U, which is the upgrade machine of EXL-50, is also presented.

physics.plasm-ph↗