Mingbin Xu
YOU?
Author Swipe
View article: Applying urinary ultrasound combined with CT to predict the risk of spontaneous ureteral stone passage
Applying urinary ultrasound combined with CT to predict the risk of spontaneous ureteral stone passage Open
Background We aimed to develop a comprehensive predictive model for spontaneous stone passage (SSP) by integrating parameters from urinary ultrasound, non-contrast computed tomography (NCCT), and clinical markers. Methods This retrospectiv…
View article: Predictors of postoperative renal function recovery in patients with ureteral stones: asymptomatic Vs symptomatic ureteral stones
Predictors of postoperative renal function recovery in patients with ureteral stones: asymptomatic Vs symptomatic ureteral stones Open
Purpose This study aimed to comprehensively compare the clinical, radiological, and functional characteristics of asymptomatic and symptomatic ureteral stones, and identify independent predictors of postoperative renal function recovery in…
View article: Prediction for spontaneous passage of ureteral stones with renal insufficiency
Prediction for spontaneous passage of ureteral stones with renal insufficiency Open
We sought to identify the predictors of spontaneous stone passage (SSP) in patients with ureteral stones, specifically those complicated by renal insufficiency and thus at greater risk for requiring intervention. Retrospective cohort study…
View article: Contextualization of ASR with LLM using phonetic retrieval-based augmentation
Contextualization of ASR with LLM using phonetic retrieval-based augmentation Open
Large language models (LLMs) have shown superb capability of modeling multimodal signals including audio and text, allowing the model to generate spoken or textual response given a speech input. However, it remains a challenge for the mode…
View article: Enhancing CTC-based speech recognition with diverse modeling units
Enhancing CTC-based speech recognition with diverse modeling units Open
In recent years, the evolution of end-to-end (E2E) automatic speech recognition (ASR) models has been remarkable, largely due to advances in deep learning architectures like transformer. On top of E2E systems, researchers have achieved sub…
View article: Conformer-Based Speech Recognition On Extreme Edge-Computing Devices
Conformer-Based Speech Recognition On Extreme Edge-Computing Devices Open
With increasingly more powerful compute capabilities and resources in today's devices, traditionally compute-intensive automatic speech recognition (ASR) has been moving from the cloud to devices to better protect user privacy. However, it…
View article: Personalization of CTC-based End-to-End Speech Recognition Using Pronunciation-Driven Subword Tokenization
Personalization of CTC-based End-to-End Speech Recognition Using Pronunciation-Driven Subword Tokenization Open
Recent advances in deep learning and automatic speech recognition have improved the accuracy of end-to-end speech recognition systems, but recognition of personal content such as contact names remains a challenge. In this work, we describe…
View article: Acoustic Model Fusion for End-to-end Speech Recognition
Acoustic Model Fusion for End-to-end Speech Recognition Open
Recent advances in deep learning and automatic speech recognition (ASR) have enabled the end-to-end (E2E) ASR system and boosted the accuracy to a new level. The E2E systems implicitly model all conventional ASR components, such as the aco…
View article: Training Large-Vocabulary Neural Language Models by Private Federated Learning for Resource-Constrained Devices
Training Large-Vocabulary Neural Language Models by Private Federated Learning for Resource-Constrained Devices Open
Federated Learning (FL) is a technique to train models using data distributed across devices. Differential Privacy (DP) provides a formal privacy guarantee for sensitive data. Our goal is to train a large neural network language model (NNL…
View article: Association between insulin-like growth factor 1 gene rs35767 polymorphisms and cancer risk
Association between insulin-like growth factor 1 gene rs35767 polymorphisms and cancer risk Open
Background: Several studies have been conducted on the relationship between insulin-like growth factor 1 gene (IGF-1) rs35767 polymorphisms and cancer risk, but the results are conflicting. We performed a meta-analysis to investigate the r…
View article: A Multi-task Learning Approach for Named Entity Recognition using Local Detection
A Multi-task Learning Approach for Named Entity Recognition using Local Detection Open
Named entity recognition (NER) systems that perform well require task-related and manually annotated datasets. However, they are expensive to develop, and are thus limited in size. As there already exists a large number of NER datasets tha…
View article: Effective Context and Fragment Feature Usage for Named Entity Recognition
Effective Context and Fragment Feature Usage for Named Entity Recognition Open
In this paper, we explore a new approach to named entity recognition (NER) with the goal of learning from context and fragment features more effectively, contributing to the improvement of overall recognition performance. We use the recent…
View article: A General FOFE-net Framework for Simple and Effective Question Answering over Knowledge Bases
A General FOFE-net Framework for Simple and Effective Question Answering over Knowledge Bases Open
Question answering over knowledge base (KB-QA) has recently become a popular research topic in NLP. One popular way to solve the KB-QA problem is to make use of a pipeline of several NLP modules, including entity discovery and linking (EDL…
View article: Fixed-Size Ordinally Forgetting Encoding Based Word Sense Disambiguation
Fixed-Size Ordinally Forgetting Encoding Based Word Sense Disambiguation Open
In this paper, we present our method of using fixed-size ordinally forgetting encoding (FOFE) to solve the word sense disambiguation (WSD) problem. FOFE enables us to encode variable-length sequence of words into a theoretically unique fix…
View article: Ethyl Pyruvate Attenuates CaCl<sub>2</sub>-Induced Tubular Epithelial Cell Injury by Inhibiting Autophagy and Inflammatory Responses
Ethyl Pyruvate Attenuates CaCl<sub>2</sub>-Induced Tubular Epithelial Cell Injury by Inhibiting Autophagy and Inflammatory Responses Open
Background/Aims: Nephrolithiasis is one of the most prevalent diseases of the urinary system. Approximately 80% of human kidney stones are composed of calcium oxalate (CaOx), and hypercalciuria is one of the most common metabolic disorders…
View article: Dual Fixed-Size Ordinally Forgetting Encoding (FOFE) for Competitive Neural Language Models
Dual Fixed-Size Ordinally Forgetting Encoding (FOFE) for Competitive Neural Language Models Open
In this paper, we propose a new approach to employ the fixed-size ordinally-forgetting encoding (FOFE) (Zhang et al., 2015b) in neural languages modelling, called dual-FOFE. The main idea of dual-FOFE is that it allows to use two different…
View article: Fixed Size Ordinally-Forgetting Encoding and its Applications
Fixed Size Ordinally-Forgetting Encoding and its Applications Open
In this thesis, we propose the new Fixed-size Ordinally-Forgetting Encoding (FOFE) method, which can almost uniquely encode any variable-length sequence of words into a fixed-size representation. FOFE can model the word order in a sequence…
View article: A Local Detection Approach for Named Entity Recognition and Mention Detection
A Local Detection Approach for Named Entity Recognition and Mention Detection Open
In this paper, we study a novel approach for named entity recognition (NER) and mention detection (MD) in natural language processing. Instead of treating NER as a sequence labeling problem, we propose a new local detection approach, which…
View article: Word Embeddings based on Fixed-Size Ordinally Forgetting Encoding
Word Embeddings based on Fixed-Size Ordinally Forgetting Encoding Open
In this paper, we propose to learn word embeddings based on the recent fixed-size ordinally forgetting encoding (FOFE) method, which can almost uniquely encode any variable-length sequence into a fixed-size representation. We use FOFE to f…
View article: A FOFE-based Local Detection Approach for Named Entity Recognition and Mention Detection
A FOFE-based Local Detection Approach for Named Entity Recognition and Mention Detection Open
In this paper, we study a novel approach for named entity recognition (NER) and mention detection in natural language processing. Instead of treating NER as a sequence labelling problem, we propose a new local detection approach, which rel…
View article: A Fixed-Size Encoding Method for Variable-Length Sequences with its Application to Neural Network Language Models
A Fixed-Size Encoding Method for Variable-Length Sequences with its Application to Neural Network Language Models Open
In this paper, we propose the new fixed-size ordinally-forgetting encoding (FOFE) method, which can almost uniquely encode any variable-length sequence of words into a fixed-size representation. FOFE can model the word order in a sequence …
View article: The Fixed-Size Ordinally-Forgetting Encoding Method for Neural Network Language Models
The Fixed-Size Ordinally-Forgetting Encoding Method for Neural Network Language Models Open
ShiLiang Zhang, Hui Jiang, MingBin Xu, JunFeng Hou, LiRong Dai. Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 2: Sh…