Search Results for author:
Found 108 papers, 42 papers with code
Syntax-guided Contrastive Learning for Pre-trained Language Model
Syntactic information has been proved to be useful for transformer-based pre-trained language models.
De-Bias for Generative Extraction in Unified NER Task
Named entity recognition (NER) is a fundamental task to recognize specific types of entities from a given sentence.
ClusterFormer: Neural Clustering Attention for Efficient and Effective Transformer
Other sparse methods use clustering patterns to select words, but the clustering process is separate from the training process of the target task, which causes a decrease in effectiveness.
Facilitating NSFW Text Detection in Open-Domain Dialogue Systems via Knowledge Distillation
NSFW (Not Safe for Work) content, in the context of a dialogue, can have severe side effects on users in open-domain dialogue systems.
InfeRE: Step-by-Step Regex Generation via Chain of Inference
Particularly, InfeRE outperforms the popular tree-based generation approach by 18. 1% and 11. 3% on both datasets, respectively, in terms of DFA@5 accuracy.
A Benchmark for Understanding Dialogue Safety in Mental Health Support
Our study reveals that ChatGPT struggles to detect safety categories with detailed safety definitions in a zero- and few-shot paradigm, whereas the fine-tuned model proves to be more suitable.
MAS: Towards Resource-Efficient Federated Multiple-Task Learning
Then, we present our new approach, MAS (Merge and Split), to optimize the performance of training multiple simultaneous FL tasks.
Latent Jailbreak: A Benchmark for Evaluating Text Safety and Output Robustness of Large Language Models
We present a systematic analysis of the safety and robustness of LLMs regarding the position of explicit normal instructions, word replacements (verbs in explicit normal instructions, target groups in malicious instructions, cue words for explicit normal instructions), and instruction replacements (different explicit normal instructions).
Combating Data Imbalances in Federated Semi-supervised Learning with Dual Regulators
Theoretically, we show the convergence guarantee of the dual regulators.
Non-Convex Optimizations for Machine Learning with Theoretical Guarantee: Robust Matrix Completion and Neural Network Learning
Despite the recent development in machine learning, most learning systems are still under the concept of "black box", where the performance cannot be understood and derived.
Understanding Client Reactions in Online Mental Health Counseling
Communication success relies heavily on reading participants' reactions.
A Semi-Paired Approach For Label-to-Image Translation
Data efficiency, or the ability to generalize from a few labeled data, remains a major challenge in deep learning.
Co-design Hardware and Algorithm for Vector Search
Vector search has emerged as the foundation for large-scale information retrieval and machine learning systems, with search engines like Google and Bing processing tens of thousands of queries per second on petabyte-scale document datasets by evaluating vector similarities between encoded query texts and web documents.
Rethinking Document-Level Relation Extraction: A Reality Check
Recently, numerous efforts have continued to push up performance boundaries of document-level relation extraction (DocRE) and have claimed significant progress in DocRE.
Patch-level Routing in Mixture-of-Experts is Provably Sample-efficient for Convolutional Neural Networks
In deep learning, mixture-of-experts (MoE) activates one or few experts (sub-networks) on a per-sample or per-token basis, resulting in significant computation reduction.
Learning Music Sequence Representation from Text Supervision
Music representation learning is notoriously difficult for its complex human-related concepts contained in the sequence of numerical signals.
Wavelet-based Unsupervised Label-to-Image Translation
Semantic Image Synthesis (SIS) is a subclass of image-to-image translation where a semantic layout is used to generate a photorealistic image.
Multimodal Unsupervised Image-To-Image Translation
Translation
+1
SMILE: Single-turn to Multi-turn Inclusive Language Expansion via ChatGPT for Mental Health Support
There has been an increasing research interest in developing specialized dialogue systems that can offer mental health support.
Prompt Pre-Training with Twenty-Thousand Classes for Open-Vocabulary Visual Recognition
This work proposes POMP, a prompt pre-training method for vision-language models.
Ranked #1 on
Open Vocabulary Semantic Segmentation
on PascalVOC-20
(hIoU metric)
MobileInst: Video Instance Segmentation on the Mobile
To address those issues, we present MobileInst, a lightweight and mobile-friendly framework for video instance segmentation on mobile devices.
Joint Edge-Model Sparse Learning is Provably Efficient for Graph Neural Networks
Due to the significant computational challenge of training large-scale graph neural networks (GNNs), various sparse learning techniques have been exploited to reduce memory and storage costs.
SKDBERT: Compressing BERT via Stochastic Knowledge Distillation
In this paper, we propose Stochastic Knowledge Distillation (SKD) to obtain compact BERT-style language model dubbed SKDBERT.
Neural Methods for Logical Reasoning Over Knowledge Graphs
We introduce a set of models that use Neural Networks to create one-point vector embeddings to answer the queries.
Profiling Television Watching Behaviour Using Bayesian Hierarchical Joint Models for Time-to-Event and Count Data
The model drastically reduces the dimensionality of the data from thousands of observations per customer to 11 customer-level parameter estimates and random effects.
Smart Multi-tenant Federated Learning
In this work, we propose a smart multi-tenant FL system, MuFL, to effectively coordinate and execute simultaneous training activities.
Learning and generalization of one-hidden-layer neural networks, going beyond standard Gaussian data
In addition, for the first time, this paper characterizes the impact of the input distributions on the sample complexity and the learning rate.
Chat-to-Design: AI Assisted Personalized Fashion Design
In this demo, we present Chat-to-Design, a new multimodal interaction system for personalized fashion design.
Optimizing Performance of Federated Person Re-identification: Benchmarking and Analysis
Based on these insights, we propose three optimization approaches: (1) We adopt knowledge distillation to facilitate the convergence of FedReID by better transferring knowledge from clients to the server; (2) We introduce client clustering to improve the performance of large datasets by aggregating clients with similar data distributions; (3) We propose cosine distance weight to elevate performance by dynamically updating the weights for aggregation depending on how well models are trained in clients.
A Fine-grained Interpretability Evaluation Benchmark for Neural NLP
Based on this benchmark, we conduct experiments on three typical models with three saliency methods, and unveil their strengths and weakness in terms of interpretability.
Modelling graph dynamics in fraud detection with "Attention"
At online retail platforms, detecting fraudulent accounts and transactions is crucial to improve customer experience, minimize loss, and avoid unauthorized transactions.
Federated Unsupervised Domain Adaptation for Face Recognition
To address this problem, we propose federated unsupervised domain adaptation for face recognition, FedFR.
Divergence-aware Federated Self-Supervised Learning
Using the framework, our study uncovers unique insights of FedSSL: 1) stop-gradient operation, previously reported to be essential, is not always necessary in FedSSL; 2) retaining local knowledge of clients in FedSSL is particularly beneficial for non-IID data.
ADD 2022: the First Audio Deep Synthesis Detection Challenge
Audio deepfake detection is an emerging topic, which was included in the ASVspoof 2021.
Reducing language context confusion for end-to-end code-switching automatic speech recognition
We propose a language-related attention mechanism to reduce multilingual context confusion for the E2E code-switching ASR model based on the Equivalence Constraint (EC) Theory.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+1
How does unlabeled data improve generalization in self-training? A one-hidden-layer theoretical analysis
Self-training, a semi-supervised learning algorithm, leverages a large amount of unlabeled data to improve learning when the labeled data are limited.
An Efficient Pruning Process with Locality Aware Exploration and Dynamic Graph Editing for Subgraph Matching
Subgraph matching is a NP-complete problem that extracts isomorphic embeddings of a query graph $q$ in a data graph $G$.
Self-Instantiated Recurrent Units with Dynamic Soft Recursion
On the other hand, the extent of the Self-IRU recursion is controlled by gates whose values are between 0 and 1 and may vary across the temporal dimension of sequences, enabling dynamic soft recursion depth at each time step.
POLLA: Enhancing the Local Structure Awareness in Long Sequence Spatial-temporal Modeling
Extensive experiments are conducted on five large-scale datasets, which demonstrate that our method achieves state-of-the-art performance and validates the effectiveness brought by local structure information.
Why Lottery Ticket Wins? A Theoretical Perspective of Sample Complexity on Pruned Neural Networks
Moreover, when the algorithm for training a pruned neural network is specified as an (accelerated) stochastic gradient descent algorithm, we theoretically show that the number of samples required for achieving zero generalization error is proportional to the number of the non-pruned weights in the hidden layer.
How unlabeled data improve generalization in self-training? A one-hidden-layer theoretical analysis
Self-training, a semi-supervised learning algorithm, leverages a large amount of unlabeled data to improve learning when the labeled data are limited.
Joint Optimization in Edge-Cloud Continuum for Federated Unsupervised Person Re-identification
We present FedUReID, a federated unsupervised person ReID system to learn person ReID models without any labels while preserving privacy.
Collaborative Unsupervised Visual Representation Learning from Decentralized Data
In this framework, each party trains models from unlabeled data independently using contrastive learning with an online network and a target network.
On Orthogonality Constraints for Transformers
Orthogonality constraints encourage matrices to be orthogonal for numerical stability.
Knowledge Router: Learning Disentangled Representations for Knowledge Graphs
To this end, this paper proposes to learn disentangled representations of KG entities - a new method that disentangles the inner latent properties of KG entities.
A Sequence-to-Set Network for Nested Named Entity Recognition
We utilize a non-autoregressive decoder to predict the final set of entities in one pass, in which we are able to capture dependencies between entities.
Ranked #6 on
Nested Named Entity Recognition
on ACE 2005
EasyFL: A Low-code Federated Learning Platform For Dummies
However, these platforms are complex to use and require a deep understanding of FL, which imposes high barriers to entry for beginners, limits the productivity of researchers, and compromises deployment efficiency.
Towards Unsupervised Domain Adaptation for Deep Face Recognition under Privacy Constraints via Federated Learning
To this end, FedFR forms an end-to-end training pipeline: (1) pre-train in the source domain; (2) predict pseudo labels by clustering in the target domain; (3) conduct domain-constrained federated learning across two domains.
Locate and Label: A Two-stage Identifier for Nested Named Entity Recognition
Although these methods have the innate ability to handle nested NER, they suffer from high computational cost, ignorance of boundary information, under-utilization of the spans that partially match with entities, and difficulties in long entity recognition.
Ranked #6 on
Nested Named Entity Recognition
on GENIA
Chinese Named Entity Recognition
named-entity-recognition
+3
FSR: Accelerating the Inference Process of Transducer-Based Models by Applying Fast-Skip Regularization
It takes a lot of computation and time to predict the blank tokens, but only the non-blank tokens will appear in the final output sequence.
TSNAT: Two-Step Non-Autoregressvie Transformer Models for Speech Recognition
To address these two problems, we propose a new model named the two-step non-autoregressive transformer(TSNAT), which improves the performance and accelerating the convergence of the NAR model by learning prior knowledge from a parameters-sharing AR model.
Beyond Fully-Connected Layers with Quaternions: Parameterization of Hypercomplex Multiplications with $1/n$ Parameters
Recent works have demonstrated reasonable success of representation learning in hypercomplex space.
Switch Spaces: Learning Product Spaces with Sparse Gating
In order for learned representations to be effective and efficient, it is ideal that the geometric inductive bias aligns well with the underlying structure of the data.
Fast End-to-End Speech Recognition via Non-Autoregressive Models and Cross-Modal Knowledge Transferring from BERT
Based on this idea, we propose a non-autoregressive speech recognition model called LASO (Listen Attentively, and Spell Once).
Learning One-hidden-layer Neural Networks on Gaussian Mixture Models with Guaranteed Generalizability
Instead of following the conventional and restrictive assumption in the literature that the input features follow the standard Gaussian distribution, this paper, for the first time, analyzes a more general and practical scenario that the input features follow a Gaussian mixture model of a finite number of Gaussian distributions of various mean and variance.
Recurrently Controlling a Recurrent Network with Recurrent Networks Controlled by More Recurrent Networks
This paper explores an intriguing idea of recursively parameterizing recurrent nets.
Why Lottery Ticket Wins? A Theoretical Perspective of Sample Complexity on Sparse Neural Networks
Moreover, as the algorithm for training a sparse neural network is specified as (accelerated) stochastic gradient descent algorithm, we theoretically show that the number of samples required for achieving zero generalization error is proportional to the number of the non-pruned model weights in the hidden layer.
Parameterization of Hypercomplex Multiplications
Recent works have demonstrated reasonable success of representation learning in hypercomplex space.
Suspicious Massive Registration Detection via Dynamic Heterogeneous Graph Neural Networks
Massive account registration has raised concerns on risk management in e-commerce companies, especially when registration increases rapidly within a short time frame.
Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting
Many real-world applications require the prediction of long sequence time-series, such as electricity consumption planning.
Ranked #5 on
Time Series Forecasting
on ETTh2 (168)
xFraud: Explainable Fraud Transaction Detection
At online retail platforms, it is crucial to actively detect the risks of transactions to improve customer experience and minimize financial loss.
Learning User Representations with Hypercuboids for Recommender Systems
Furthermore, we present two variants of hypercuboids to enhance the capability in capturing the diversities of user interests.
Decoupling Pronunciation and Language for End-to-end Code-switching Automatic Speech Recognition
In this paper, we propose a decoupled transformer model to use monolingual paired data and unpaired text data to alleviate the problem of code-switching data shortage.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+1
One In A Hundred: Select The Best Predicted Sequence from Numerous Candidates for Streaming Speech Recognition
Inspired by the success of two-pass end-to-end models, we introduce a transformer decoder and the two-stage inference method into the streaming CTC model.
MicroRec: Efficient Recommendation Inference by Hardware and Data Structure Solutions
MicroRec accelerates recommendation inference by (1) redesigning the data structures involved in the embeddings to reduce the number of lookups needed and (2) taking advantage of the availability of High-Bandwidth Memory (HBM) in FPGA accelerators to tackle the latency by enabling parallel lookups.
Improving Network Slimming with Nonconvex Regularization
Network slimming with T$\ell_1$ regularization also outperforms the latest Bayesian modification of network slimming in compressing a CNN architecture in terms of memory storage while preserving its model accuracy after channel pruning.
Clustering COVID-19 Lung Scans
While tests do exist for COVID-19, the goal of our research is to explore other methods of identifying infected individuals.
Computational prediction of RNA tertiary structures using machine learning methods
RNAs play crucial and versatile roles in biological processes.
A Practical Chinese Dependency Parser Based on A Large-scale Dataset
DDParser is extended on the graph-based biaffine parser to accommodate to the characteristics of Chinese dataset.
Exploring particle dynamics during self-organization processes via rotationally invariant latent representations
The dynamic of complex ordering systems with active rotational degrees of freedom exemplified by protein self-assembly is explored using a machine learning workflow that combines deep learning-based semantic segmentation and rotationally invariant variational autoencoder-based analysis of orientation and shape evolution.
Soft Condensed Matter
Performance Optimization for Federated Person Re-identification via Benchmark Analysis
Then we propose two optimization methods: (1) To address the unbalanced weight problem, we propose a new method to dynamically change the weights according to the scale of model changes in clients in each training round; (2) To facilitate convergence, we adopt knowledge distillation to refine the server model with knowledge generated from client models on a public dataset.
TensorCoder: Dimension-Wise Attention via Tensor Representation for Natural Language Modeling
Transformer has been widely-used in many Natural Language Processing (NLP) tasks and the scaled dot-product attention between tokens is a core module of Transformer.
Fast Learning of Graph Neural Networks with Guaranteed Generalizability: One-hidden-layer Case
In this paper, we provide a theoretically-grounded generalizability analysis of GNNs with one hidden layer for both regression and binary classification problems.
Automated Radiological Report Generation For Chest X-Rays With Weakly-Supervised End-to-End Deep Learning
The model provides automated recognition of given scans and generation of reports.
Spike-Triggered Non-Autoregressive Transformer for End-to-End Speech Recognition
To address this problem and improve the inference speed, we propose a spike-triggered non-autoregressive transformer model for end-to-end speech recognition, which introduces a CTC module to predict the length of the target sequence and accelerate the convergence.
Listen Attentively, and Spell Once: Whole Sentence Generation via a Non-Autoregressive Architecture for Low-Latency Speech Recognition
Without beam-search, the one-pass propagation much reduces inference time cost of LASO.
TRP: Trained Rank Pruning for Efficient Deep Neural Networks
The TRP trained network inherently has a low-rank structure, and is approximated with negligible performance loss, thus eliminating the fine-tuning process after low rank decomposition.
Rnn-transducer with language bias for end-to-end Mandarin-English code-switching speech recognition
Recently, language identity information has been utilized to improve the performance of end-to-end code-switching (CS) speech recognition.
$\ell_0$ Regularized Structured Sparsity Convolutional Neural Networks
Deepening and widening convolutional neural networks (CNNs) significantly increases the number of trainable weight parameters by adding more convolutional layers and feature maps per layer, respectively.
Synchronous Transformers for End-to-End Speech Recognition
Once a fixed-length chunk of the input sequence is processed by the encoder, the decoder begins to predict symbols immediately.
Integrating Knowledge into End-to-End Speech Recognition from External Text-Only Data
To alleviate the above two issues, we propose a unified method called LST (Learn Spelling from Teachers) to integrate knowledge into an AED model from the external text-only data and leverage the whole context in a sentence.
Language Modelling
Sequence-To-Sequence Speech Recognition
+1
Weakly-Supervised Degree of Eye-Closeness Estimation
To alleviate this issue, we propose a weakly-supervised method which utilizes the accurate annotation from the synthetic data set, to learn accurate degree of eye openness, and the weakly labeled (open or closed) real world eye data set to control the domain shift.
DeGNN: Characterizing and Improving Graph Neural Networks with Graph Decomposition
Despite the wide application of Graph Convolutional Network (GCN), one major limitation is that it does not benefit from the increasing depth and suffers from the oversmoothing problem.
Trained Rank Pruning for Efficient Deep Neural Networks
To accelerate DNNs inference, low-rank approximation has been widely adopted because of its solid theoretical rationale and efficient implementations.
R2D2: Reuse & Reduce via Dynamic Weight Diffusion for Training Efficient NLP Models
We propose R2D2 layers, a new neural block for training efficient NLP models.
Holographic Factorization Machines for Recommendation
Factorization Machines (FMs) are a class of popular algorithms that have been widely adopted for collaborative filtering and recommendation tasks.
A Tensorized Transformer for Language Modeling
In this paper, based on the ideas of tensor decomposition and parameters sharing, we propose a novel self-attention model (namely Multi-linear attention) with Block-Term Tensor Decomposition (BTD).
Fully Decoupled Neural Network Learning Using Delayed Gradients
Training neural networks with back-propagation (BP) requires a sequential passing of activations and gradients, which forces the network modules to work in a synchronous fashion.
Lightweight and Efficient Neural Natural Language Processing with Quaternion Networks
Many state-of-the-art neural models for NLP are heavily parameterized and thus memory inefficient.
Quaternion Collaborative Filtering for Recommendation
All in all, we conduct extensive experiments on six real-world datasets, demonstrating the effectiveness of Quaternion algebra in recommender systems.
DeepRec: An Open-source Toolkit for Deep Learning based Recommendation
In this toolkit, we have implemented a number of deep learning based recommendation algorithms using Python and the widely used deep learning package - Tensorflow.
Quaternion Knowledge Graph Embeddings
In this work, we move beyond the traditional complex-valued representations, introducing more expressive hypercomplex representations to model entities and relations for knowledge graph embeddings.
Ranked #4 on
Link Prediction
on FB15k
Understanding Straight-Through Estimator in Training Activation Quantized Neural Nets
We prove that if the STE is properly chosen, the expected coarse gradient correlates positively with the population gradient (not available for the training), and its negation is a descent direction for minimizing the population loss.
AutoShuffleNet: Learning Permutation Matrices via an Exact Lipschitz Continuous Penalty in Deep Convolutional Neural Networks
In addition, we found experimentally that the standard convex relaxation of permutation matrices into stochastic matrices leads to poor performance.
DAC: Data-free Automatic Acceleration of Convolutional Networks
A complex deep learning model with high accuracy runs slowly on resource-limited devices, while a light-weight model that runs much faster loses accuracy.
Trained Rank Pruning for Efficient Deep Neural Networks
We propose Trained Rank Pruning (TRP), which iterates low rank approximation and training.
DNQ: Dynamic Network Quantization
Network quantization is an effective method for the deployment of neural networks on memory and energy constrained mobile devices.
HyperML: A Boosting Metric Learning Approach in Hyperbolic Space for Recommender Systems
This paper investigates the notion of learning user and item representations in non-Euclidean space.
Ranked #1 on
Recommendation Systems
on MovieLens 20M
(HR@10 metric)
Next Item Recommendation with Self-Attention
In this paper, we propose a novel sequence-aware recommendation model.
Blended Coarse Gradient Descent for Full Quantization of Deep Neural Networks
We introduce the notion of coarse gradient and propose the blended coarse gradient descent (BCGD) algorithm, for training fully quantized neural networks.
GrCAN: Gradient Boost Convolutional Autoencoder with Neural Decision Forest
We develop a gradient boost module and embed it into the proposed convolutional autoencoder with neural decision forest to improve the performance.
Self-Attentive Neural Collaborative Filtering
This paper has been withdrawn as we discovered a bug in our tensorflow implementation that involved accidental mixing of vectors across batches.
NeuRec: On Nonlinear Transformation for Personalized Ranking
Modeling user-item interaction patterns is an important task for personalized recommendations.
Metric Factorization: Recommendation beyond Matrix Factorization
In the past decade, matrix factorization has been extensively researched and has become one of the most popular techniques for personalized recommendations.
BinaryRelax: A Relaxation Approach For Training Deep Neural Networks With Quantized Weights
We propose BinaryRelax, a simple two-phase algorithm, for training deep neural networks with quantized weights.
Stacked Kernel Network
Similar to DNNs, a SKN is composed of multiple layers of hidden units, but each parameterized by a RKHS function rather than a finite-dimensional vector.
Deep Learning based Recommender System: A Survey and New Perspectives
This article aims to provide a comprehensive review of recent research efforts on deep learning based recommender systems.
Quantization and Training of Low Bit-Width Convolutional Neural Networks for Object Detection
We present LBW-Net, an efficient optimization based method for quantization and training of the low bit-width convolutional neural networks (CNNs).
Mimicing the Kane-Mele type spin orbit interaction by spin-flexual phonon coupling in graphene devices
On the efforts of enhancing the spin orbit interaction (SOI) of graphene for seeking the dissipationless quantum spin Hall devices, unique Kane-Mele type SOI and high mobility samples are desired.
Mesoscale and Nanoscale Physics
