Search Results for author:
Found 96 papers, 26 papers with code
End-to-end Speech Translation System Description of LIT for IWSLT 2019
We propose layer-tied self-attention for end-to-end speech translation.
Federated Data Model
In artificial intelligence (AI), especially deep learning, data diversity and volume play a pivotal role in model development.
Spatiotemporal Diffusion Model with Paired Sampling for Accelerated Cardiac Cine MRI
Current deep learning reconstruction for accelerated cardiac cine MRI suffers from spatial and temporal blurring.
Clinically Feasible Diffusion Reconstruction for Highly-Accelerated Cardiac Cine MRI
The currently limited quality of accelerated cardiac cine reconstruction may potentially be improved by the emerging diffusion models, but the clinically unacceptable long processing time poses a challenge.
GenNBV: Generalizable Next-Best-View Policy for Active 3D Reconstruction
Previous works attempt to automate this process using the Next-Best-View (NBV) policy for active 3D reconstruction.
Triplet-constraint Transformer with Multi-scale Refinement for Dose Prediction in Radiotherapy
Radiotherapy is a primary treatment for cancers with the aim of applying sufficient radiation dose to the planning target volume (PTV) while minimizing dose hazards to the organs at risk (OARs).
EmbodiedScan: A Holistic Multi-Modal 3D Perception Suite Towards Embodied AI
In the realm of computer vision and robotics, embodied agents are expected to explore their environment and carry out human instructions.
Spectroscopy-Guided Discovery of Three-Dimensional Structures of Disordered Materials with Diffusion Models
In this work, we introduce a new framework based on the diffusion model, a recent generative machine learning method to predict 3D structures of disordered materials from a target property.
CAMEL2: Enhancing weakly supervised learning for histopathology images by incorporating the significance ratio
Histopathology image analysis plays a crucial role in cancer diagnosis.
Improving End-to-End Speech Processing by Efficient Text Data Utilization with Latent Synthesis
For SLU, LaSyn improves our E2E baseline by absolute 4. 1% for intent classification accuracy and 3. 8% for slot filling SLU-F1 on SLURP, and absolute 4. 49% and 2. 25% for exact match (EM) and EM-Tree accuracies on STOP respectively.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+6
Enhancing Cardiac MRI Segmentation via Classifier-Guided Two-Stage Network and All-Slice Information Fusion Transformer
Cardiac Magnetic Resonance imaging (CMR) is the gold standard for assessing cardiac function.
DiffusePast: Diffusion-based Generative Replay for Class Incremental Semantic Segmentation
The Class Incremental Semantic Segmentation (CISS) extends the traditional segmentation task by incrementally learning newly added classes.
Chance-Constrained Multiple-Choice Knapsack Problem: Model, Algorithms, and Applications
In particular, we focus on the practical scenario of CCMCKP, where the probability distributions of random weights are unknown but only sample data is available.
Fairly Adaptive Negative Sampling for Recommendations
Pairwise learning strategies are prevalent for optimizing recommendation models on implicit feedback data, which usually learns user preference by discriminating between positive (i. e., clicked by a user) and negative items (i. e., obtained by negative sampling).
An Unsupervised Framework for Joint MRI Super Resolution and Gibbs Artifact Removal
Previous studies tackled these two problems separately, where super resolution methods tend to enhance Gibbs artifacts, whereas Gibbs ringing removal methods tend to blur the images.
Computationally Efficient 3D MRI Reconstruction with Adaptive MLP
Recon3DMLP improves HR 3D reconstruction and outperforms several existing CNN-based models under similar GPU memory consumption, which demonstrates that Recon3DMLP is a practical solution for HR 3D MRI reconstruction.
Holistic Multi-Slice Framework for Dynamic Simultaneous Multi-Slice MRI Reconstruction
Deep learning (DL)-based methods have shown promising results for single-slice MR reconstruction, but the addition of SMS acceleration raises unique challenges due to the composite k-space signals and the resulting images with strong inter-slice artifacts.
M4Singer: a Multi-Style, Multi-Singer and Musical Score Provided Mandarin Singing Corpus
The lack of publicly available high-quality and accurately labeled datasets has long been a major bottleneck for singing voice synthesis (SVS).
JoJoNet: Joint-contrast and Joint-sampling-and-reconstruction Network for Multi-contrast MRI
The proposed framework consists of a sampling mask generator for each image contrast and a reconstructor exploiting the inter-contrast correlations with a recurrent structure which enables the information sharing in a holistic way.
A Comprehensive Survey on Trustworthy Recommender Systems
As one of the most successful AI-powered applications, recommender systems aim to help people make appropriate decisions in an effective and efficient way, by providing personalized suggestions in many aspects of our lives, especially for various human-oriented online services such as e-commerce platforms and social media sites.
Automatic Comment Generation via Multi-Pass Deliberation
Then, we treat the comment of the retrieved code as the initial draft and input it with the code and keywords into DECOM to start the iterative deliberation process.
Deep Statistic Shape Model for Myocardium Segmentation
Additionally, the predicted point cloud guarantees boundary correspondence for sequential images, which contributes to the downstream tasks, such as the motion estimation of myocardium.
Robust Landmark-based Stent Tracking in X-ray Fluoroscopy
Even though angioplasty devices are designed to have radiopaque markers for the ease of tracking, current methods struggle to deliver satisfactory results due to the small marker size and complex scenes in angioplasty.
Low-bit Shift Network for End-to-End Spoken Language Understanding
Deep neural networks (DNN) have achieved impressive success in multiple domains.
Invertible Sharpening Network for MRI Reconstruction Enhancement
During inference, the learned blurring transform can be inverted to a sharpening transform leveraging the network's invertibility.
DProQ: A Gated-Graph Transformer for Protein Complex Structure Assessment
We challenge this significant task with DProQ, which introduces a gated neighborhood-modulating Graph Transformer (GGT) designed to predict the quality of 3D protein complex structures.
EGR: Equivariant Graph Refinement and Assessment of 3D Protein Complex Structures
Protein complexes are macromolecules essential to the functioning and well-being of all living organisms.
DeepFD: Automated Fault Diagnosis and Localization for Deep Learning Programs
Besides, for fault localization, DeepFD also outperforms the existing works, correctly locating 42% faulty programs, which almost doubles the best result (23%) achieved by the existing works.
Constructing dynamic residential energy lifestyles using Latent Dirichlet Allocation
In this research, we propose and test a new framework for understanding residential electricity demand by using a dynamic energy lifestyles approach that is iterative and highly extensible.
CorrectSpeech: A Fully Automated System for Speech Correction and Accent Reduction
This study propose a fully automated system for speech correction and accent reduction.
Accelerating Representation Learning with View-Consistent Dynamics in Data-Efficient Reinforcement Learning
Learning informative representations from image-based observations is of fundamental concern in deep Reinforcement Learning (RL).
CoCA-MDD: A Coupled Cross-Attention based Framework for Streaming Mispronunciation Detection and Diagnosis
We utilize conv-transformer structure to encode input speech in a streaming manner.
An Adaptive-Importance-Sampling-Enhanced Bayesian Approach for Topology Estimation in an Unbalanced Power Distribution System
To overcome the above challenges, this paper proposes a Bayesian-inference framework that allows us to simultaneously estimate the topology and the state of a three-phase, unbalanced power distribution system.
bert2BERT: Towards Reusable Pretrained Language Models
However, large language model pre-training costs intensive computational resources and most of the models are trained from scratch without reusing the existing pre-trained models, which is wasteful.
Reinforcement Learning with Predictive Consistent Representations
Learning informative representations from image-based observations is a fundamental problem in deep Reinforcement Learning (RL).
Efficient, Interpretable Graph Neural Network Representation for Angle-dependent Properties and its Application to Optical Spectroscopy
In this work, we extend the recently proposed ALIGNN encoding, which incorporates bond angles, to also include dihedral angles (ALIGNN-d).
DyLex: Incorporating Dynamic Lexicons into BERT for Sequence Labeling
Incorporating lexical knowledge into deep learning models has been proved to be very effective for sequence labeling tasks.
GhostBERT: Generate More Features with Cheap Operations for BERT
Transformer-based pre-trained language models like BERT, though powerful in many tasks, are expensive in both memory and computation, due to their large number of parameters.
AutoTinyBERT: Automatic Hyper-parameter Optimization for Efficient Pre-trained Language Models
Specifically, we carefully design the techniques of one-shot learning and the search space to provide an adaptive and efficient development way of tiny PLMs for various latency constraints.
EditSpeech: A Text Based Speech Editing System Using Partial Inference and Bidirectional Fusion
This paper presents the design, implementation and evaluation of a speech editing system, named EditSpeech, which allows a user to perform deletion, insertion and replacement of words in a given speech utterance, without causing audible degradation in speech quality and naturalness.
Membraneless organelles formed by liquid-liquid phase separation increase bacterial fitness
Liquid-liquid phase separation is emerging as a crucial phenomenon in several fundamental cell processes.
VQMIVC: Vector Quantization and Mutual Information-Based Unsupervised Speech Representation Disentanglement for One-shot Voice Conversion
One-shot voice conversion (VC), which performs conversion across arbitrary speakers with only a single target-speaker utterance for reference, can be effectively achieved by speech representation disentanglement.
Unsupervised Domain Adaptation for Dysarthric Speech Detection via Domain Adversarial Training and Mutual Information Minimization
Such systems are particularly susceptible to domain mismatch where the training and testing data come from the source and target domains respectively, but the two domains may differ in terms of speech stimuli, disease etiology, etc.
Multi-scale Neural ODEs for 3D Medical Image Registration
The inference consists of iterative gradient updates similar to a conventional gradient descent optimizer but in a much faster way, because the neural ODE learns from the training data to adapt the gradient efficiently at each iteration.
Learning about latent dynamic trading demand
We show that trading strategies are combinations of trading towards investor targets, liquidity provision for other investors' demands, and front-running based on learning about latent underlying trading demand imbalances and future price pressure.
Automatic Construction of Sememe Knowledge Bases via Dictionaries
We use this method to build an English SKB and a French SKB, and conduct comprehensive evaluations from both intrinsic and extrinsic perspectives.
A Streaming End-to-End Framework For Spoken Language Understanding
In this paper, we propose a streaming end-to-end framework that can process multiple intentions in an online and incremental way.
Cardiac Functional Analysis with Cine MRI via Deep Learning Reconstruction
To the best of our knowledge, this is the first work to evaluate the cine MRI with deep learning reconstruction for cardiac function analysis and compare it with other conventional methods.
Accelerating 3D MULTIPLEX MRI Reconstruction with Deep Learning
We propose a deep learning framework for undersampled 3D MRI data reconstruction and apply it to MULTIPLEX MRI.
Extract then Distill: Efficient and Effective Task-Agnostic BERT Distillation
Task-agnostic knowledge distillation, a teacher-student framework, has been proved effective for BERT compression.
Non-unitary Entanglement Dynamics in Continuous Variable Systems
We construct a random unitary Gaussian circuit for continuous-variable (CV) systems subject to Gaussian measurements.
Quantum Physics Disordered Systems and Neural Networks Statistical Mechanics
LightMBERT: A Simple Yet Effective Method for Multilingual BERT Distillation
The multilingual pre-trained language models (e. g, mBERT, XLM and XLM-R) have shown impressive performance on cross-lingual natural language understanding tasks.
Improving Task-Agnostic BERT Distillation with Layer Mapping Search
Comprehensive experiments on the evaluation benchmarks demonstrate that 1) layer mapping strategy has a significant effect on task-agnostic BERT distillation and different layer mappings can result in quite different performances; 2) the optimal layer mapping strategy from the proposed search process consistently outperforms the other heuristic ones; 3) with the optimal layer mapping, our student model achieves state-of-the-art performance on the GLUE tasks.
Generating private data with user customization
Personal devices such as mobile phones can produce and store large amounts of data that can enhance machine learning models; however, this data may contain private information specific to the data owner that prevents the release of the data.
Group-Skeleton-Based Human Action Recognition in Complex Events
In this paper, we propose a novel group-skeleton-based human action recognition method in complex events.
HyperText: Endowing FastText with Hyperbolic Geometry
Natural language data exhibit tree-like hierarchical structures such as the hypernym-hyponym relations in WordNet.
The HUAWEI Speaker Diarisation System for the VoxCeleb Speaker Diarisation Challenge
This paper describes system setup of our submission to speaker diarisation track (Track 4) of VoxCeleb Speaker Recognition Challenge 2020.
TernaryBERT: Distillation-aware Ultra-low Bit BERT
Transformer-based pre-training models like BERT have achieved remarkable performance in many natural language processing tasks. However, these models are both computation and memory expensive, hindering their deployment to resource-constrained devices.
Anatomy-Aware Cardiac Motion Estimation
A baseline dense motion tracker is trained to approximate the motion fields and then refined to estimate anatomy-aware motion fields under the weak supervision from the VAE.
Conv-Transformer Transducer: Low Latency, Low Frame Rate, Streamable End-to-End Speech Recognition
To reduce computation cost, we gradually downsample acoustic input, also with the interleaved convolution layers.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+1
Real-Time Cardiac Cine MRI with Residual Convolutional Recurrent Neural Network
Real-time cardiac cine MRI does not require ECG gating in the data acquisition and is more useful for patients who can not hold their breaths or have abnormal heart rhythms.
Aspect Sentiment Classification with Document-level Sentiment Preference Modeling
This justifies the importance of the document-level sentiment preference information to ASC and the effectiveness of our approach capturing such information.
Motion Pyramid Networks for Accurate and Efficient Cardiac Motion Estimation
In this paper, we propose Motion Pyramid Networks, a novel deep learning-based approach for accurate and efficient cardiac motion estimation.
Effective Data Fusion with Generalized Vegetation Index: Evidence from Land Cover Segmentation in Agriculture
How can we effectively leverage the domain knowledge from remote sensing to better segment agriculture land cover from satellite images?
The 1st Agriculture-Vision Challenge: Methods and Results
The first Agriculture-Vision Challenge aims to encourage research in developing novel and effective algorithms for agricultural pattern recognition from aerial images, especially for the semantic segmentation task associated with our challenge dataset.
Leaderless Consensus of a Hierarchical Cyber-Physical System
This paper models a class of hierarchical cyber-physical systems and studies its associated consensus problem.
DynaBERT: Dynamic BERT with Adaptive Width and Depth
The pre-trained language models like BERT, though powerful in many natural language processing tasks, are both computation and memory expensive.
Computational Performance of a Germline Variant Calling Pipeline for Next Generation Sequencing
With the booming of next generation sequencing technology and its implementation in clinical practice and life science research, the need for faster and more efficient data analysis methods becomes pressing in the field of sequencing.
FOAL: Fast Online Adaptive Learning for Cardiac Motion Estimation
In clinical deployment, however, they suffer dramatic performance drops due to mismatched distributions between training and testing datasets, commonly encountered in the clinical environment.
Advanced Variations of Two-Dimensional Principal Component Analysis for Face Recognition
To enhance the generalization ability of extracted features, a novel relaxed 2DPCA (R2DPCA) is proposed with a new ridge regression model.
Pyramid Convolutional RNN for MRI Image Reconstruction
We evaluate our model on the fastMRI knee and brain datasets and the results show that the proposed model outperforms other methods and can recover more details.
A General Framework for Adaptation of Neural Machine Translation to Simultaneous Translation
Despite the success of neural machine translation (NMT), simultaneous neural machine translation (SNMT), the task of translating in real time before a full sentence has been observed, remains challenging due to the syntactic structure difference and simultaneity requirements.
Energy Resource Control via Privacy Preserving Data
Although the frequent monitoring of smart meters enables granular control over energy resources, it also increases the risk of leakage of private information such as income, home occupancy, and power consumption behavior that can be inferred from the data by an adversary.
Systems and Control Computers and Society Systems and Control
TinyBERT: Distilling BERT for Natural Language Understanding
To accelerate inference and reduce model size while maintaining accuracy, we first propose a novel Transformer distillation method that is specially designed for knowledge distillation (KD) of the Transformer-based models.
Ranked #1 on
Natural Language Inference
on MultiNLI Dev
NEZHA: Neural Contextualized Representation for Chinese Language Understanding
The pre-trained language models have achieved great successes in various natural language understanding (NLU) tasks due to its capacity to capture the deep contextualized information in text by pre-training on large-scale corpora.
Dialog State Tracking with Reinforced Data Augmentation
Neural dialog state trackers are generally limited due to the lack of quantity and diversity of annotated training data.
Understanding Video Content: Efficient Hero Detection and Recognition for the Game "Honor of Kings"
In order to understand content and automatically extract labels for videos of the game "Honor of Kings", it is necessary to detect and recognize characters (called "hero") together with their camps in the game video.
News Cover Assessment via Multi-task Learning
We propose an end-to-end multi-task learning network for image clarity assessment and semantic segmentation simultaneously, the results of which can be guided for news cover assessment.
Modeling Semantic Compositionality with Sememe Knowledge
In this paper, we verify the effectiveness of sememes, the minimum semantic units of human languages, in modeling SC by a confirmatory experiment.
multi-word expression embedding
multi-word expression sememe prediction
Relaxed 2-D Principal Component Analysis by $L_p$ Norm for Face Recognition
A relaxed two dimensional principal component analysis (R2DPCA) approach is proposed for face recognition.
Generative Adversarial Models for Learning Private and Fair Representations
We present Generative Adversarial Privacy and Fairness (GAPF), a data-driven framework for learning private and fair representations of the data.
Distributed generation of privacy preserving data with user customization
Distributed devices such as mobile phones can produce and store large amounts of data that can enhance machine learning models; however, this data may contain private information specific to the data owner that prevents the release of the data.
Measurement-driven entanglement transition in hybrid quantum circuits
In all models we find a volume law entangled phase for low measurement rates, which exhibits a sub-dominant logarithmic behavior in the entanglement entropy, S_A = {\alpha} ln |A| + s|A|, with sub-system size |A|.
Statistical Mechanics Disordered Systems and Neural Networks Strongly Correlated Electrons Quantum Physics
An Investigation of Few-Shot Learning in Spoken Term Classification
In this paper, we investigate the feasibility of applying few-shot learning algorithms to a speech task.
Data-Rate Driven Transmission Strategy for Deep Learning Based Communication Systems
The GDR scheme can achieve higher data rate than the conventional one-hot vector scheme with comparable BLER performance.
Information Theory Information Theory
Brain Segmentation from k-space with End-to-end Recurrent Attention Network
The task of medical image segmentation commonly involves an image reconstruction step to convert acquired raw data to images before any analysis.
Understanding Compressive Adversarial Privacy
Designing a data sharing mechanism without sacrificing too much privacy can be considered as a game between data holders and malicious attackers.
Dual-label Deep LSTM Dereverberation For Speaker Verification
In this paper, we present a reverberation removal approach for speaker verification, utilizing dual-label deep neural networks (DNNs).
Learning Efficient Single-stage Pedestrian Detectors by Asymptotic Localization Fitting
However, current single-stage detectors (e. g. SSD) have not presented competitive accuracy on common pedestrian detection benchmarks.
Ranked #11 on
Pedestrian Detection
on Caltech
(using extra training data)
Android HIV: A Study of Repackaging Malware for Evading Machine-Learning Detection
In contrast to existing works, the adversarial examples crafted by our method can also deceive recent machine learning based detectors that rely on semantic features such as control-flow-graph.
Cryptography and Security
Generative Adversarial Privacy
We present a data-driven framework called generative adversarial privacy (GAP).
Infrastructure Quality Assessment in Africa using Satellite Imagery and Deep Learning
Our best models predict infrastructure quality with AUROC scores of 0. 881 on Electricity, 0. 862 on Sewerage, 0. 739 on Piped Water, and 0. 786 on Roads using Landsat 8.
Context-Aware Generative Adversarial Privacy
On the one hand, context-free privacy solutions, such as differential privacy, provide strong privacy guarantees, but often lead to a significant reduction in utility.
Statistical Parametric Speech Synthesis Using Generative Adversarial Networks Under A Multi-task Learning Framework
In this paper, we aim at improving the performance of synthesized speech in statistical parametric speech synthesis (SPSS) based on a generative adversarial network (GAN).
Sound
A propagation matting method based on the Local Sampling and KNN Classification with adaptive feature space
Second, adaptively use local sampling or using local KNN classifier for processing based on the pros and cons of the sample performance of unknown image areas.
