A5102852549- Polynomial and algebraic computation
- Mechanical and Optical Resonators
- Photonic and Optical Devices
- Formal Methods in Verification
- Photonic Crystal and Fiber Optics
- Advanced Fiber Optic Sensors
- Advanced Optimization Algorithms Research
- Advanced Numerical Analysis Techniques
- Software Testing and Debugging Techniques
- Generative Adversarial Networks and Image Synthesis
- Sparse and Compressive Sensing Techniques
- Numerical Methods and Algorithms
- Digital Filter Design and Implementation
- Advanced Fiber Laser Technologies
- Software Engineering Research
- Logic, programming, and type systems
- Antenna Design and Optimization
- Force Microscopy Techniques and Applications
- Strong Light-Matter Interactions
- Plant Parasitism and Resistance
- Commutative Algebra and Its Applications
- Software Reliability and Analysis Research
- Computational Geometry and Mesh Generation
- Software System Performance and Reliability
- Quantum Information and Cryptography
Huawei Technologies (China)
2024
The Audio Recording Academy
2024
Wuhan University
2024
Peking University
2012-2023
BOE Technology Group (China)
2023
Ludong University
2023
Beijing Institute of Technology
2023
China Academy of Engineering Physics
2022
South China University of Technology
2020-2022
Peng Cheng Laboratory
2022
Coherent light-matter interaction at the single photon and electronic qubit level promises remarkable potential for nonclassical information processing. Against efforts of improving figure merit cavities, here we demonstrate strong anharmonicity in polariton dressed states via dark state resonances a highly dissipative cavity. It is shown that vacuum Rabi oscillation occurs quantum emitter inside cavity even with bosonic decay-to-interaction rate ratio exceeding $10^{2}$, when field coupled...
The research in AI-based formal mathematical reasoning has shown an unstoppable growth trend. These studies have excelled competitions like IMO, showing significant progress. However, these intertwined multiple skills simultaneously, i.e., problem-solving, reasoning, and writing specifications, making it hard to precisely identify the LLMs' strengths weaknesses each task. This paper focuses on verification, immediate application scenario of decomposes into six sub-tasks. We constructed 18k...
We propose to realize effective beam-splitter-like and two-mode-squeezing photon-photon interactions in a strong coupling optomechanical interface by exploiting detuned driving lasers. In this interface, the transitions between optical system mechanical oscillator are suppressed large energy offsets, therefore protecting from dissipations. Moreover, destructive quantum interference eigenmodes of is capable further reducing effects initial thermal occupations. The can serve as universal block...
We investigate the single-photon transport in a single-mode optical fiber coupled to an optomechanical system strong-coupling regime. The transmission amplitude is analytically obtained with real-space approach and effects of cavity mechanical dissipations are studied via master-equation simulations. Based on theoretical framework, we further propose heralded probabilistic scheme generate NOON states arbitrary phonon numbers by measuring sideband photons. efficiency fidelity discussed finally.
We propose a realistic system with separated optical and mechanical degrees of freedom, in which high-mechanical-quality silicon nitride membrane is placed upon high-optical-quality whispering gallery microcavity. The strongly enhanced linear optomechanical coupling, together simultaneously low losses the present system, results remarkable single-photon cooperativity exceeding 300. This unprecedented enables nonlinearity at light intensities holds great potential generating, storing,...
A new sparse SOS decomposition algorithm is proposed based on a sparsity pattern, called cross patterns. The pattern focuses the of terms and thus different from well-known correlative which variables though algorithms these two patterns both take use chordal extensions/chordal decompositions. Moreover, it proved that obtained by always refinement block-diagonalization sign-symmetry method. %Because covers more polynomials than Various experiments show dramatically saves computational cost...
We theoretically investigate optomechanical coupling between a single-wall carbon nanotube and high-$Q$ whispering gallery microcavity. The anisotropy in the refractive index of results polarization-sensitive coupling. Moreover, band structure leads to strong dependence strength on wavelength. In particular, pure dissipative can be achieved at absorption resonance frequency nanotube, single-photon exceed $10\phantom{\rule{0.28em}{0ex}}\text{kHz}$. This induces cooling oscillation,...
From 1929, Edwin Link created a flight simulator, to today's PICO4 virtual reality all-in-one machine. the first attempt at technology, now China's domestic PICO company has developed new VR This is undoubtedly huge step forward. In technology industry, in addition company, world also Meta's brand Oculus, Valve Corporation and others leading development of industry. Oculus largest market. Here, by comparing product research different companies, this paper studies industry today gives...
We address the problem of facial motion retargeting that aims to transfer from a 2D face image 3D characters. Existing methods often formulate this as reconstruction problem, which estimates attributes such identity and expression images. However, due lack ground-truth labels for both expression, most 3D-face reconstruction-based fail capture accurately. As result, these may not achieve promising performance. To this, we propose an identity-consistent constraint learn accurate identities by...
Cylindrical algebraic decomposition (CAD) plays an important role in the field of real geometry and many other areas. As is well-known, choice variable ordering while computing CAD has a great effect on time memory use computation as well number sample points computed. In this paper, we indicate that typical algorithms, if executed with respect to special kind orderings (called "the perfect elimination orderings''), naturally preserve chordality, which compatible (variable) sparsity pattern...
In the domain of large-scale software development, demands for dynamic and multifaceted static code analysis exceed capabilities traditional tools. To bridge this gap, we present CodeFuse-Query, a system that redefines through fusion Domain Optimized System Design Logic Oriented Computation Design. CodeFuse-Query reimagines as data computation task, support scanning over 10 billion lines daily more than 300 different tasks. It optimizes resource utilization, prioritizes reusability, applies...
Formal verification provides a rigorous and systematic approach to ensure the correctness reliability of software systems. Yet, constructing specifications for full proof relies on domain expertise non-trivial manpower. In view such needs, an automated specification synthesis is desired. While existing approaches are limited in their versatility, i.e., they either focus only synthesizing loop invariants numerical programs, or tailored specific types programs invariants. Programs involving...
Abstract In the last few years, rapid advancement of Internet Things (IoT) and widespread adoption smart cities have posed new challenges to computing services. Traditional cloud models fail fulfil response requirement latency‐sensitive applications, while mobile edge (MEC) improves service efficiency customer experience by transferring tasks servers located at network edge. However, designing an effective offloading strategy in complex scenarios involving multiple tasks, nodes, services...
Motivation: Try to obtaining high spatial and temporal resolution image in dynamic MRI. Goal(s): The plug play truncated low rank optimization combined iterative soft thresholding(PNPT-) embedded method was proposed reconstructe data tested accurately estimate the function for dMRI reconstruction. Approach: PNPT- uses nuclear norm thresholding assign different shrink values singular vectors, enabling preservation of essential information while effectively eliminating noise. Results: Impact:...
Neural network quantization is an effective way to compress deep models and improve their execution latency energy efficiency, so that they can be deployed on mobile or embedded devices. Existing methods require original data for calibration fine-tuning get better performance. However, in many real-world scenarios, the may not available due confidential private issues, thereby making existing applicable. Moreover, absence of data, recently developed generative adversarial networks (GANs)...
Mixed trigonometric-polynomials (MTPs) are functions of the form $f(x,\sin{x}, \cos{x})$ with $f\in\mathbb{Q}[x_1,x_2,x_3]$. In this paper, an algorithm ``isolating" all real roots MTP is provided and implemented. It automatically divides into two parts: one consists finitely many ``bounded" in interval $[\mu_-,\mu_+]$ while other probably countably ``periodic" $\mathbb{R}\backslash[\mu_-,\mu_+]$. For bounded roots, returns isolating intervals corresponding multiplicities for periodic it...