Search Results for author: Jun Kong

Found 9 papers, 1 papers with code

Accelerating Inference for Pretrained Language Models by Unified Multi-Perspective Early Exiting

no code implementations COLING 2022 Jun Kong, Jin Wang, Liang-Chih Yu, Xuejie Zhang

To address this limitation, a unified horizontal and vertical multi-perspective early exiting (MPEE) framework is proposed in this study to accelerate the inference of transformer-based models.

Multi-scale frequency separation network for image deblurring

no code implementations1 Jun 2022 Yanni Zhang, Qiang Li, Miao Qi, Di Liu, Jun Kong, Jianzhong Wang

MSFS-Net introduces the frequency separation module (FSM) into an encoder-decoder network architecture to capture the low- and high-frequency information of image at multiple scales.

Contrastive Learning Deblurring +1

Last-iterate convergence analysis of stochastic momentum methods for neural networks

no code implementations30 May 2022 Dongpo Xu, Jinlan Liu, Yinghua Lu, Jun Kong, Danilo Mandic

The stochastic momentum method is a commonly used acceleration technique for solving large-scale stochastic optimization problems in artificial neural networks.

Stochastic Optimization

Image deblurring based on lightweight multi-information fusion network

no code implementations14 Jan 2021 Yanni Zhang, Yiming Liu, Qiang Li, Miao Qi, Dahong Xu, Jun Kong, Jianzhong Wang

In the encoding stage, the image feature is reduced to various smallscale spaces for multi-scale information extraction and fusion without a large amount of information loss.

Deblurring Image Deblurring

HPCC-YNU at SemEval-2020 Task 9: A Bilingual Vector Gating Mechanism for Sentiment Analysis of Code-Mixed Text

1 code implementation SEMEVAL 2020 Jun Kong, Jin Wang, Xuejie Zhang

In this paper, we (my codalab username is kongjun) present a system that uses a bilingual vector gating mechanism for bilingual resources to complete the task.

Sentiment Analysis

Clumped Nuclei Segmentation with Adjacent Point Match and Local Shape based Intensity Analysis for Overlapped Nuclei in Fluorescence In-Situ Hybridization Images

no code implementations14 Aug 2018 Xiaoyuan Guo, Hanyi Yu, Blair Rossetti, George Teodoro, Daniel Brat, Jun Kong

Highly clumped nuclei clusters captured in fluorescence in situ hybridization microscopy images are common histology entities under investigations in a wide spectrum of tissue-related biomedical investigations.

Analysis of Cellular Feature Differences of Astrocytomas with Distinct Mutational Profiles Using Digitized Histopathology Images

no code implementations24 Jun 2018 Mousumi Roy, Fusheng Wang, George Teodoro, Jose Velazqeuz Vega, Daniel Brat, Jun Kong

Cellular phenotypic features derived from histopathology images are the basis of pathologic diagnosis and are thought to be related to underlying molecular profiles.

Dimensionality Reduction

Cannot find the paper you are looking for? You can Submit a new open access paper.