
GSPMD: General and Scalable Parallelization for ML Computation Graphs
We present GSPMD, an automatic, compilerbased parallelization system fo...
read it

Improving Named Entity Recognition by External Context Retrieving and Cooperative Learning
Recent advances in Named Entity Recognition (NER) show that documentlev...
read it

Quantum Synchronizable Codes on Sextic Cyclotomy
Quantum synchronizable codes are kinds of quantum errorcorrecting codes...
read it

FedCom: A ByzantineRobust Local Model Aggregation Rule Using Data Commitment for Federated Learning
Federated learning (FL) is a promising privacypreserving distributed ma...
read it

HalfTruth: A Partially Fake Audio Detection Dataset
Diverse promising datasets have been designed to hold back the developme...
read it

IDOLNet: An Interactive DualDomain Parallel Network for CT Metal Artifact Reduction
Due to the presence of metallic implants, the imaging quality of compute...
read it

Auto Correcting in the Process of Translation – Multitask Learning Improves Dialogue Machine Translation
Automatic translation of dialogue texts is a much needed demand in many ...
read it

A Universal Model for Cross Modality Mapping by Relational Reasoning
With the aim of matching a pair of instances from two different modaliti...
read it

Regional and Sectoral Structures and Their Dynamics of Chinese Economy: A Network Perspective from MultiRegional InputOutput Tables
A multiregional inputoutput table (MRIOT) containing the transactions ...
read it

Attention Models for Point Clouds in Deep Learning: A Survey
Recently, the advancement of 3D point clouds in deep learning has attrac...
read it

DANNet: DualDomain AdaptiveScaling Nonlocal Network for CT Metal Artifact Reduction
Metal implants can heavily attenuate Xrays in computed tomography (CT) ...
read it

Isolation mechanisms for highspeed packetprocessing pipelines
Dataplane programmability is now mainstream, both in the form of progra...
read it

TokenstoToken ViT: Training Vision Transformers from Scratch on ImageNet
Transformers, which are popular for language modeling, have been explore...
read it

MobilityAware Seamless Handover with MPTCP in SoftwareDefined HetNets
In this paper, the problem of vertical handover in softwaredefined netw...
read it

An Investigation of Potential Function Designs for Neural CRF
The neural linearchain CRF model is one of the most widelyused approac...
read it

Exploring the limits of Concurrency in ML Training on Google TPUs
Recent results in language understanding using neural networks have requ...
read it

You Recommend, I Buy: How and Why People Engage in Instant Messaging Based Social Commerce
As an emerging business phenomenon especially in China, instant messagin...
read it

EDCNN: Edge enhancementbased Densely Connected Network with Compound Loss for LowDose CT Denoising
In the past few decades, to reduce the risk of Xray in computed tomogra...
read it

Star edgecoloring of some special graphs
The star chromatic index of a multigraph G, denoted by χ_star'(G), is th...
read it

Understanding the Role of Intermediaries in Online Social Ecommerce: An Exploratory Study of Beidian
Social ecommerce, as a new form of social computing based marketing pla...
read it

Toward Accurate Personlevel Action Recognition in Videos of Crowded Scenes
Detecting and recognizing human action in videos with crowded scenes is ...
read it

Rapid Robust Principal Component Analysis: CUR Accelerated Inexact Low Rank Estimation
Robust principal component analysis (RPCA) is a widely used tool for dim...
read it

Structural Knowledge Distillation
Knowledge distillation is a critical technique to transfer knowledge bet...
read it

Automated Concatenation of Embeddings for Structured Prediction
Pretrained contextualized embeddings are powerful word representations f...
read it

More Embeddings, Better Sequence Labelers?
Recent work proposes a family of contextual embeddings that significantl...
read it

Fast and Accurate Sequence Labeling with Approximate Inference Network
The linearchain Conditional Random Field (CRF) model is one of the most...
read it

Finding Action Tubes with a SparsetoDense Framework
The task of spatialtemporal action detection has attracted increasing a...
read it

The Devil is in Classification: A Simple Framework for Longtail Instance Segmentation
Most existing object instance detection and segmentation models only wor...
read it

A MultiLevel Approach to Waste Object Segmentation
We address the problem of localizing waste objects from a color image an...
read it

Overcoming Classifier Imbalance for Longtail Object Detection with Balanced Group Softmax
Solving longtail large vocabulary object detection with deep learning b...
read it

Double Circulant Selfdual Codes on Sextic Cyclotomy
This paper contributes to construct double circulant selfdual codes by ...
read it

Gauntlet: Finding Bugs in Compilers for Programmable Packet Processing
Programmable packetprocessing devices such as programmable switches and...
read it

Making Robots Draw A Vivid Portrait In Two Minutes
Significant progress has been made with artistic robots. However, existi...
read it

Human in Events: A LargeScale Benchmark for Humancentric Video Analysis in Complex Events
Along with the development of the modern smart city, humancentric video...
read it

Automatic lowbit hybrid quantization of neural networks through meta learning
Model quantization is a widely used technique to compress and accelerate...
read it

StructureLevel Knowledge Distillation For Multilingual Sequence Labeling
Multilingual sequence labeling is a task of predicting label sequences u...
read it

Discontinuous Galerkin method for a distributed optimal control problem governed by a time fractional diffusion equation
This paper is devoted to the numerical analysis of a control constrained...
read it

A multiple attributes image quality database for smartphone camera photo quality assessment
Smartphone is the superstar product in digital device market and the qua...
read it

Numerical analysis of two Galerkin discretizations with graded temporal grids for fractional evolution equations
Two numerical methods with graded temporal grids are analyzed for fracti...
read it

LargeScale Discrete Fourier Transform on TPUs
In this work, we present two parallel algorithms for the largescale dis...
read it

CTM: Collaborative Temporal Modeling for Action Recognition
With the rapid development of digital multimedia, video understanding ha...
read it

iqiyi Submission to ActivityNet Challenge 2019 Kinetics700 challenge: Hierarchical Groupwise Attention
In this report, the method for the iqiyi submission to the task of Activ...
read it

Learning a Layout Transfer Network for Context Aware Object Detection
We present a context aware object detection method based on a retrievea...
read it

Merging External Bilingual Pairs into Neural Machine Translation
As neural machine translation (NMT) is not easily amenable to explicit c...
read it

Classification Calibration for Longtail Instance Segmentation
Remarkable progress has been made in object instance detection and segme...
read it

Revisit Knowledge Distillation: a Teacherfree Framework
Knowledge Distillation (KD) aims to distill the knowledge of a cumbersom...
read it

Scale MLPerf0.6 models on Google TPUv3 Pods
The recent submission of Google TPUv3 Pods to the industry wide MLPerf ...
read it

DP4coloring of planar graphs with some restrictions on cycles
DPcoloring was introduced by Dvořák and Postle as a generalization of l...
read it

Open Named Entity Modeling from Embedding Distribution
In this paper, we report our discovery on named entity distribution in g...
read it

Numerical analysis of a semilinear fractional diffusion equation
This paper considers the numerical analysis of a semilinear fractional d...
read it
Tao Wang
is this you? claim profile
MuTao Wang is a Taiwanese mathematician and current professor of mathematics at the University of Columbia.
In 1984, originally for international business, he entered the National University of Taiwan, and after a year he switched to mathematics. He received his B.S. in Mathematics from the National University of Taiwan in 1988 and his M.S. degree from the same institution in 1992. His thesis “Generalized harmonic maps and representations of discrete groups” was awarded by a PhD in Mathematics at Harvard University in 1998. His thesis counselor at Harvard was the Chinese Fields Medalistic and the ShingTung Yau differential geometer.
Wang became an Associate Professor at the Columbia Faculty in 2001 and a full Professor in 2009. Wang was an assistant professor at Stanford University before joining the faculty in Columbia. From 2003–2005 he was a Sloan Research Fellow. In 2007 he was awarded the Chern Prize as the Kavli Fellow of the National Academy of Sciences. In 2010, Wang was a Plenary Speaker in the International Congress of Chinese Mathematicians at the International Congress of Chinese Mathematicians and a Plenary Speaker in the International Congress of mathematical physics at the International Congress on Mathematical Physics. In addition, he also spoke plenary at the International Conference on Differential Geometry in 2011. After winning the Morningside Medal, Wang told interviewers that he wasn’t a very good student and didn’t consistently grade well. It has struggled to study topics that did not only interest him for the grade, but spends a lot of time on topics that interest him. He credits his mathematical career to two persons: his mother and his thesis consultant ShingTung Yau. He cites the support of his mother and understands her decision to change to Mathematics in universities, despite being a considerably less lucrative field, and describes Yau as the pivotal point of his life in 1992 when he decided to focus primarily upon mathematics research.
Wang’s research focuses on differential geometry and mathematical physics and on general relativity in particular. He studied extensively the greater codimensional flow of the mean curvature, leading to criteria for existence, regularity and convergence of the flow. In the field of general relativity, he is known especially for his work on quasilocal mass energy; in his honour, the nearlocal mass of WangYau is named.