Min Zhang's research while affiliated with University of Pittsburgh and other places

Publications (538)

Article
Recently, character information has been successfully introduced into the encoder-decoder event detection model to relieve the trigger-word mismatch problem, thus achieving impressive results in the languages without natural delimiters (i.e., Chinese). However, it is introduced into the encoder or the decoder separately, which makes the advantage o...
Preprint
Full-text available
Visual Entailment with natural language explanations aims to infer the relationship between a text-image pair and generate a sentence to explain the decision-making process. Previous methods rely mainly on a pre-trained vision-language model to perform the relation inference and a language model to generate the corresponding explanation. However, t...
Conference Paper
Full-text available
Recently, pre-training methods tailored for IR tasks have achieved great success. However, as the mechanisms behind the performance improvement remain under-investigated, the interpretability and robustness of these pre-trained models still need to be improved. Axiomatic IR aims to identify a set of desirable properties expressed mathematically as...
Article
Full-text available
Generating questions from the long passage is an important and challenging task. Most of the recent works focus on generating questions whose answers are consecutive text spans in the given passage. However, realistic questions are more complicated and their answers are always inductive and summative. In this paper, we focus on a complex form of qu...
Conference Paper
In the past few years, cross-modal image-text retrieval (ITR) has experienced increased interest in the research community due to its excellent research value and broad real-world application. It is designed for the scenarios where the queries are from one modality and the retrieval galleries from another modality. This paper presents a comprehensi...
Conference Paper
Most previous studies of document-level event extraction mainly focus on building argument chains in an autoregressive way, which achieves a certain success but is inefficient in both training and inference. In contrast to the previous studies, we propose a fast and lightweight model named as PTPCG. In our model, we design a novel strategy for even...
Conference Paper
Document-level neural machine translation (DocNMT) universally encodes several local sentences or the entire document. Thus, DocNMT does not consider the relevance of document-level contextual information, for example, some context (i.e., content words, logical order, and co-occurrence relation) is more effective than another auxiliary context (i.e...
Article
Document ranking is one of the most studied but challenging problems in information retrieval (IR). More and more studies have begun to address this problem from fine-grained document modeling. However, most of them focus on context-independent passage-level relevance signals and ignore the context information. In this paper, we investigate how inf...
Article
Pre-training and fine-tuning have become the de facto paradigm in many natural language processing (NLP) tasks. However, compared to other NLP tasks, neural machine translation (NMT) aims to generate target language sentences through the contextual representation from the source language counterparts. This characteristic means the optimization obje...
Article
Implicit discourse relation recognition (IDRR) is a challenging but crucial task in discourse analysis. Most existing methods train multiple models to predict multi-level labels independently, while ignoring the dependence between hierarchically structured labels. In this paper, we consider multi-level IDRR as a conditional label sequence generatio...
Preprint
Full-text available
Medical dialogue generation is an important yet challenging task. Most previous works rely on the attention mechanism and large-scale pretrained language models. However, these methods often fail to acquire pivotal information from the long dialogue history to yield an accurate and informative response, due to the fact that the medical entities usu...
Article
Full-text available
Cancer treatments such as chemotherapies may change or accelerate aging trajectories in cancer patients. Emerging evidence has shown that “omics” data can be used to study molecular changes of the aging process. Here, we integrated the drug-induced and normal aging transcriptomic data to computationally characterize the potential cancer drug-induce...
Preprint
During the past decade, neural network models have made tremendous progress on in-domain semantic role labeling (SRL). However, performance drops dramatically under the out-of-domain setting. In order to facilitate research on cross-domain SRL, this paper presents MuCPAD, a multi-domain Chinese predicate-argument dataset, which consists of 30,897 s...
Preprint
Self-augmentation has been received increasing research interest recently to improve named entity recognition (NER) performance in low-resource scenarios. Token substitution and mixup are two feasible heterogeneous self-augmentation techniques for NER that can achieve effective performance with certain specialized efforts. Noticeably, self-augmenta...
Preprint
This paper presents MuCGEC, a multi-reference multi-source evaluation dataset for Chinese Grammatical Error Correction (CGEC), consisting of 7,063 sentences collected from three different Chinese-as-a-Second-Language (CSL) learner sources. Each sentence has been corrected by three annotators, and their corrections are meticulously reviewed by an ex...
Preprint
Recent works of opinion expression identification (OEI) rely heavily on the quality and scale of the manually-constructed training corpus, which could be extremely difficult to satisfy. Crowdsourcing is one practical solution for this problem, aiming to create a large-scale but quality-unguaranteed corpus. In this work, we investigate Chinese OEI w...
Preprint
Full-text available
Non-autoregressive (NAR) generation, which is first proposed in neural machine translation (NMT) to speed up inference, has attracted much attention in both machine learning and natural language processing communities. While NAR generation can significantly accelerate inference speed for machine translation, the speedup comes at the cost of sacrifi...
Preprint
Full-text available
Data augmentations (DA) are the cores to achieving robust sequence-to-sequence learning on various natural language processing (NLP) tasks. However, most of the DA approaches force the decoder to make predictions conditioned on the perturbed input representation, underutilizing supervised information provided by perturbed input. In this work, we pr...
Article
Full-text available
Background: Atrial natriuretic peptide (ANP) has been associated with cardiovascular disease (CVD) and related risk factors, but the clinical application is limited and the underlying mechanisms are not very clear. Here, we aimed to examine whether proANP and its coding gene methylation were associated with CVD in the Chinese population. Methods:...
Article
Background As a key enzyme of the natriuretic peptides system, corin may participate in the development of cardiovascular disease (CVD). Its level in circulation predicted CVD recurrence in patients with myocardial infarction and heart failure, but no study examined this prediction in general populations. Objectives This study sought to examine th...
Article
How to effectively model global context has been a critical challenge for document-level neural machine translation (NMT). Both preceding and global context have been carefully explored in the sequence-to-sequence (seq2seq) framework. However, previous studies generally map global context into one vector, which is not enough to well represent the e...
Preprint
Full-text available
In the past few years, cross-modal image-text retrieval (ITR) has experienced increased interest in the research community due to its excellent research value and broad real-world application. It is designed for the scenarios where the queries are from one modality and the retrieval galleries from another modality. This paper presents a comprehensi...
Preprint
In recent years, large-scale pre-trained language models (PLMs) have made extraordinary progress in most NLP tasks. But, in the unsupervised POS tagging task, works utilizing PLMs are few and fail to achieve state-of-the-art (SOTA) performance. The recent SOTA performance is yielded by a Guassian HMM variant proposed by He et al. (2018). However, a...
Preprint
Residual networks are an Euler discretization of solutions to Ordinary Differential Equations (ODE). This paper explores a deeper relationship between Transformer and numerical ODE methods. We first show that a residual block of layers in Transformer can be described as a higher-order solution to ODE. Inspired by this, we design a new architecture,...
Article
Background: Furin has been associated with diabetes but the underlying mechanisms are unclear. As a mediator linking fixed genome and dynamic environment, DNA methylation of its coding gene FURIN may be involved. Here, we aimed to examine the prospective association between DNA methylation in FURIN promoter and incident diabetes during 4 years of...
Preprint
Most dominant neural machine translation (NMT) models are restricted to make predictions only according to the local context of preceding words in a left-to-right manner. Although many previous studies try to incorporate global information into NMT models, there still exist limitations on how to effectively exploit bidirectional global context. In...
Article
Full-text available
Background Furin has been associated with glucose metabolic phenotypes in small sampled clinical studies. However, this association has not yet been studied in Chinese. Here, we aimed to examine the association between serum furin and fasting glucose in Chinese adults. Methods Serum furin and fasting plasma glucose were assayed for 2,172 participa...
Article
One of the important factors affecting micro-video recommender systems is to model the multi-modal user preference on the micro-video. Despite the remarkable performance of prior arts, they are still limited by fusing the user preference derived from different modalities in a unified manner, ignoring the users tend to place different emphasis on di...
Preprint
Implicit discourse relation recognition (IDRR) is a challenging but crucial task in discourse analysis. Most existing methods train multiple models to predict multi-level labels independently, while ignoring the dependence between hierarchically structured labels. In this paper, we consider multi-level IDRR as a conditional label sequence generatio...
Preprint
Text-based person search aims to retrieve images of a certain pedestrian by a textual description. The key challenge of this task is to eliminate the inter-modality gap and achieve the feature alignment across modalities. In this paper, we propose a semantic-aligned embedding method for text-based person search, in which the feature alignment acros...
Preprint
Sequential recommendation methods play an important role in real-world recommender systems. These systems are able to catch user preferences by taking advantage of historical records and then performing recommendations. Contrastive learning(CL) is a cutting-edge technology that can assist us in obtaining informative user representations, but these...
Preprint
Full-text available
There are two main challenges in document-level event extraction: 1) argument entities are scattered in different sentences, and 2) event triggers are often not available. To address these challenges, most previous studies mainly focus on building argument chains in an autoregressive way, which is inefficient in both training and inference. In cont...
Article
Full-text available
In this study, we report a novel role of metabotropic glutamate receptor 4 (GRM4) in suppressing antitumor immunity. We revealed in three murine syngeneic tumor models (B16, MC38, and 3LL) that either genetic knockout (Grm4−/−) or pharmacological inhibition led to significant delay in tumor growth. Mechanistically, perturbation of GRM4 resulted in...
Preprint
Currently, BIO-based and tuple-based approaches perform quite well on the span-based semantic role labeling (SRL) task. However, the BIO-based approach usually needs to encode a sentence once for each predicate when predicting its arguments, and the tuple-based approach has to deal with a huge search space of $O(n^3)$, greatly reducing the training...
Article
Sentiment classification aims to identify the polarity of a given review. Most existing methods consider each review as an individual while ignoring the importance of the user and product information of the given review. A direct way to integrate user and product information is to employ an attention mechanism to learn the local interaction between...
Article
Through the development of large-scale natural language models with writing and dialogue capabilities, artificial intelligence (AI) has taken a significant stride towards better natural language understanding (NLU) and human-computer interaction (HCI). As of today, the GPT-3 model, developed by OpenAI, is the language model with the most parameters...
Article
TikTok is one of the world's most attractive and fastest-growing social media platforms. It has more than 2.6 billion downloads globally and over 100 million users in the US. The “secret weapon” is its unique methodology of discovering and delivering content. On the “For You” feed, TikTok mixes videos of both internet celebrities and newcomers, enc...
Article
Existing multi-turn context-response matching methods mainly concentrate on obtaining multi-level and multi-dimension representations and better interactions between context utterances and response. However, in real-place conversation scenarios, whether a response candidate is suitable not only counts on the given dialogue context but also other ba...
Article
Existing studies for multi-source neural machine translation (NMT) either separately model different source sentences or resort to the conventional single-source NMT by simply concatenating all source sentences. However, there exist two drawbacks in these approaches. First, they ignore the explicit word-level semantic interactions between source se...
Preprint
Semantic role labeling is a fundamental yet challenging task in the NLP community. Recent works of SRL mainly fall into two lines:1) BIO-based and 2) span-based. Despite effectiveness, they share some intrinsic drawbacks of not explicitly considering internal argument structures, which may potentially hinder the model's expressiveness. To remedy th...
Article
Most of the deep neural machine translation (NMT) models are based on a bottom-up feedforward fashion, in which representations in low layers construct or modulate high layers representations. We conjecture that this unidirectional encoding fashion could be a potential issue in building a deep NMT model. In this paper, we propose to build a deeper...
Chapter
Emotion classification has become a hot research topic in natural language processing due to its wide application. Existing studies suffer from the error propagation problem when using the syntax information in emotion classification since the parser can not produce perfect syntax trees. To address this problem, we propose a new approach by compari...
Conference Paper
Conversational discourse structures aim to describe how a dialogue is organized, thus they are helpful for dialogue understanding and response generation. This paper focuses on predicting discourse dependency structures for multi-party dialogues. Previous work adopts incremental methods that take the features from the already predicted discourse re...
Article
Most state-of-the-art neural machine translation (NMT) models progressively encode feature representation in a bottom-up feed-forward fashion. This traditional encoding mechanism has no guidance from external signals. In computer vision tasks, the feedback connection plays a crucial role, particularly for understanding tasks. In this paper, we prop...
Conference Paper
Document context-aware machine translation remains challenging due to the lack of large-scale document parallel corpora. To make full use of source-side monolingual documents for context-aware NMT, we propose a Pre-training approach with Global Context (PGC). In particular, we first propose a novel self-supervised pre-training task, which contains...
Preprint
Recent studies on Knowledge Base Question Answering (KBQA) have shown great progress on this task via better question understanding. Previous works for encoding questions mainly focus on the word sequences, but seldom consider the information from syntactic trees.In this paper, we propose an approach to learn syntax-based representations for KBQA....
Article
The cardiac conduction system is a network of heterogeneous cell population that initiates and propagates electric excitations in the myocardium. Purkinje fibers, a network of specialized myocardial cells, comprise the distal end of the conduction system in the ventricles. The developmental origins of Purkinje fibers and their roles during cardiac...
Preprint
Dropout is a powerful and widely used technique to regularize the training of deep neural networks. In this paper, we introduce a simple regularization strategy upon dropout in model training, namely R-Drop, which forces the output distributions of different sub models generated by dropout to be consistent with each other. Specifically, for each tr...
Preprint
A well-known limitation in pretrain-finetune paradigm lies in its inflexibility caused by the one-size-fits-all vocabulary. This potentially weakens the effect when applying pretrained models into natural language generation (NLG) tasks, especially for the subword distributions between upstream and downstream tasks with significant discrepancy. Tow...
Preprint
Full-text available
Bilingual Lexicon Induction (BLI) aims to map words in one language to their translations in another, and are typically through learning linear projections to align monolingual word representation spaces. Two classes of word representations have been explored for BLI: static word embeddings and contextual representations, but there is no studies to...
Preprint
Full-text available
Unlike English letters, Chinese characters have rich and specific meanings. Usually, the meaning of a word can be derived from its constituent characters in some way. Several previous works on syntactic parsing propose to annotate shallow word-internal structures for better utilizing character-level information. This work proposes to model the deep...
Preprint
Full-text available
Self-attention based models have achieved remarkable success in natural language processing. However, the self-attention network design is questioned as suboptimal in recent studies, due to its veiled validity and high redundancy. In this paper, we focus on pre-trained language models with self-pruning training design on task-specific tuning. We de...
Preprint
Existing multi-turn context-response matching methods mainly concentrate on obtaining multi-level and multi-dimension representations and better interactions between context utterances and response. However, in real-place conversation scenarios, whether a response candidate is suitable not only counts on the given dialogue context but also other ba...
Preprint
Full-text available
Data augmentation has attracted a lot of research attention in the deep learning era for its ability in alleviating data sparseness. The lack of data for unseen evaluation databases is exactly the major challenge for cross-domain text-to-SQL parsing. Previous works either require human intervention to guarantee the quality of generated data \cite{y...
Article
Dependency syntax has been demonstrated highly useful for a number of natural language processing (NLP) tasks. Typical approaches of utilizing dependency syntax include Tree-RNN and Tree-Linearization, both of which exploit explicit 1-best tree outputs from a well-trained parser as inputs. However, these approaches may suffer from error propagation...