Tian Shi

ORCID: 0009-0003-6363-3585
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Topic Modeling
  • Nonlinear Dynamics and Pattern Formation
  • Service-Oriented Architecture and Web Services
  • Microwave Dielectric Ceramics Synthesis
  • Natural Language Processing Techniques
  • Neural Networks Stability and Synchronization
  • Advanced Computational Techniques and Applications
  • X-ray Diffraction in Crystallography
  • Text and Document Classification Technologies
  • Solid-state spectroscopy and crystallography
  • Ferroelectric and Piezoelectric Materials
  • Nuclear materials and radiation effects
  • Advanced ceramic materials synthesis
  • Metal and Thin Film Mechanics
  • Advanced Neural Network Applications
  • Multimodal Machine Learning Applications
  • Surface Modification and Superhydrophobicity
  • Magnesium Oxide Properties and Applications
  • Industrial Vision Systems and Defect Detection
  • Geomechanics and Mining Engineering
  • Slime Mold and Myxomycetes Research
  • Diamond and Carbon-based Materials Research
  • Electromagnetic wave absorption materials
  • Millimeter-Wave Propagation and Modeling
  • Cognitive Computing and Networks

Guilin University of Electronic Technology
2023

Nanjing University of Finance and Economics
2023

Hefei University of Technology
2019

Wuhan University of Technology
2013

In this paper, a nematic liquid crystal (NLC)-based tunable terahertz (THz) plasmonic metamaterials (MMs) with large modulation depth (MD) and low insertion loss (IL) is designed experimentally verified at THz frequencies. The proposed structure includes two-layered MM that immersed in LC. metal used directly as electrode. device 46×46 array of sub-wavelength circular air loops was fabricated on quartz glass substrate, 2×2 cm2 area 220 µm thickness. obtained results show the amplitude MD IL...

10.1364/oe.27.027039 article EN cc-by Optics Express 2019-09-11

10.1016/j.physa.2023.128571 article EN Physica A Statistical Mechanics and its Applications 2023-02-16

10.1109/bibm62325.2024.10822485 article EN 2021 IEEE International Conference on Bioinformatics and Biomedicine (BIBM) 2024-12-03

This article investigates a higher-order dynamical network on simplicial complexes, and the evolution of topological signals residing simplices (e.g. nodes, links, triangles, etc) is governed by coupled higher-dimensional linear oscillators. We uncover stability conditions cluster synchronization both its projected systems corresponding to solenoidal irrotational component dynamics, phenomena can co-exist (i.e., symbiotic synchronization). In framework this study indicates that there are...

10.1109/tnse.2023.3321140 article EN IEEE Transactions on Network Science and Engineering 2023-10-02

SSD is a single-stage target detection algorithm, which performs feature extraction by convolutional neural network and takes different layers for output, so multi-scale method. In the layer to be detected, 3*3 convolution directly used perform transformation of channels. ssd uses an anchor strategy with pre-defined anchors aspect ratios, each output predicts multiple frames (4 or 6) based on anchor. A approach used, where shallow detect small targets deep large targets. yolov5 adds some new...

10.25236/ajcis.2023.060611 article EN Academic Journal of Computing & Information Science 2023-01-01

In recent times, sequence-to-sequence (seq2seq) models have gained a lot of popularity and provide state-of-the-art performance in wide variety tasks such as machine translation, headline generation, text summarization, speech to conversion, image caption generation. The underlying framework for all these is usually deep neural network comprising an encoder decoder. Although simple encoder-decoder produce competitive results, many researchers proposed additional improvements over models,...

10.48550/arxiv.1805.09461 preprint EN other-oa arXiv (Cornell University) 2018-01-01

Event Detection (ED) is an important task in natural language processing. In the past few years, many datasets have been introduced for advancing ED machine learning models. However, most of these are under-explored because not tools available people to study events, trigger words, and event mention instances systematically efficiently. this paper, we present interactive easy-to-use tool, namely Explorer, dataset model exploration. Explorer consists web application, API, NLP toolkit, which...

10.48550/arxiv.2204.12456 preprint EN other-oa arXiv (Cornell University) 2022-01-01

Constructing comprehensive performance-enhanced heterostructured composite by using the synergistic effect of complementary advantages between heterogeneous materials has always been a challenging topic. In this work, highly stable β-NaYF4:Yb3+, Tm3+@BiOCl (NYF@BiOCl) upconversion composites were successfully prepared two-step hydrothermal method. Compared to NYF/BiOCl mechanical mixtures, NYF@BiOCl with intimate interface show higher photocatalytic degradation rate rhodamine B (RhB) aqueous...

10.2139/ssrn.4303285 article EN SSRN Electronic Journal 2022-01-01
Coming Soon ...