Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks232 from public data
- Words Can Shift: Dynamically Adjusting Word Representations Using Nonverbal Behaviors494
The Recurrent Attended Variation Embedding Network (RAVEN) is proposed that models the fine-grained structure of nonverbal subword sequences and dynamically shifts word representations based on nonverbal cues to capture the dynamic nature of non verbal intents.
- A Survey on Recent Advances in LLM-Based Multi-turn Dialogue Systems269
This article aims to elaborate recent advances in multi-turn dialogue systems, covering both LLM-based open-domain dialogue (ODD) and task-oriented dialogue (TOD) systems, along with datasets and evaluation metrics.
- A Comprehensive Survey on Relation Extraction: Recent Advances and New Frontiers164
This survey provides a comprehensive review of existing deep learning techniques for Relation extraction and proposes a new taxonomy to categorize existing works from three perspectives, i.e., text representation, context encoding, and triplet prediction.
- 163
- 152
- A Survey of Natural Language Generation149
This survey aims to give the latest synthesis of deep learning research on the NLG core tasks, as well as the architectures adopted in the field, and detail meticulously and comprehensively various NLG tasks and datasets.
- Natural Language Understanding and Inference with MLLM in Visual Question Answering: A Survey138
This survey gives an up-to-date synthesis of natural language understanding of images and text, as well as the knowledge reasoning module based on image-question information on the core VQA tasks.
- Time-driven feature-aware jointly deep reinforcement learning for financial signal representation and algorithmic trading123
A time-driven feature-aware jointly deep reinforcement learning model (TFJ-DRL) that integrates deep learning model and reinforcementLearning model to improve the financial signal representation learning and action decision making in algorithmic trading is proposed.
- Generalized Adjusted Rand Indices for cluster ensembles107
Adjusted Rand Index is generalized to two new measures based on matrix comparison: (i) Adjusted Rand Index between a similarity matrix and a cluster partition (ARImp), to evaluate the consistency of a set of clustering solutions with their corresponding consensus matrix in a cluster ensemble, and (ii) AdjustedRand Index between similarity matrices (ARImm), to evaluation the consistency between two similarity matrix.
- Chinese Relation Extraction with Multi-Grained Information and External Linguistic Knowledge96
A multi-grained lattice framework for Chinese relation extraction is proposed, which incorporates word-level information into character sequence inputs so that segmentation errors can be avoided and model multiple senses of polysemous words with the help of external linguistic knowledge to alleviate polysemy ambiguity.
- An ontology-driven clinical decision support system (IDDAP) for infectious disease diagnosis and antibiotic prescription86
This study attempted to bridge the patient/caregiver gap by building a sophisticated application that uses artificial intelligence and machine learning computational techniques to perform data-driven decision-making at the point of primary care.
- 80
- Psychometric Properties of Patient-Reported Outcome Measures for Use in Patients with Anterior Cruciate Ligament Injuries79
The ACL-RSI is a reliable and valid scale to measure outcomes for patients with ACL injuries and further assessments with improved methodological quality are needed to determine the psychometric properties of other instruments.
- Emerging medical informatics with case-based reasoning for aiding clinical decision in multi-agent system78
This research aims to depict the methodological steps and tools about the combined operation of case-based reasoning (CBR) and multi-agent system (MAS) to expose the ontological application in the field of clinical decision support.
- 72
- Knowledge-aware Attentive Neural Network for Ranking Question Answer Pairs70
A context-knowledge interactive learning architecture is developed, in which a context-guided attentive convolutional neural network is designed to integrate knowledge embeddings into sentence representations and a knowledge-aware attention mechanism is presented to attend interrelations between each segments of QA pairs.
- Investigating the transferring capability of capsule networks for text classification68
The transferring capability of the proposed cross-domain capsule network (TL-Capsule) is demonstrated by applying it to two transfer learning applications: single-label to multi-label text classification and cross- domain sentiment classification.
- 66
- Joint Learning of Answer Selection and Answer Summary Generation in Community Question Answering65
A question-driven pointer-generator network is designed, which exploits the correlation information between question-answer pairs to aid in attending the essential information when generating answer summaries and a new large-scale CQA corpus is constructed, WikiHowQA, which contains long answers for answer selection as well as reference summaries for answer summarization.
- 65
- FM-ECG: A fine-grained multi-label framework for ECG image classification60
Experimental results on two real-world large-scale ECG datasets prove the capability of FM-ECG comparing with other state-of-the-art methods in ECG abnormally detection, and visualization analyses on attention parts show that meaningful spatial attention can be effectively learned by FM- ECG.
- 59
- 59
- Multi-Task Learning with Multi-View Attention for Answer Selection and Knowledge Base Question Answering55
A novel multi-task learning scheme that utilizes multi-view attention learned from various perspectives to enable these tasks to interact with each other as well as learn more comprehensive sentence representations is proposed.
- 54
- Feature-enhanced attention network for target-dependent sentiment classification52
A Feature-enhanced Attention Network to improve the performance of target-dependent Sentiment classification (FANS) is proposed and the experimental results demonstrate that FANS has robust superiority over competitors and sets state-of-the-art.
- 47
- A Survey on Mechanistic Interpretability for Multi-Modal Foundation Models43
A structured taxonomy of interpretability methods is proposed, insights are compared across unimodal and multimodal architectures, and critical research gaps are highlighted to highlight critical research gaps.
- Continual Learning for Task-oriented Dialogue System with Iterative Network Pruning, Expanding and Masking43
This paper proposes an effective continual learning method for the task-oriented dialogue system with iterative network pruning, expanding, and masking (TPEM), which preserves performance on previously encountered tasks while accelerating learning progress on subsequent tasks.
- Multitask Learning and Reinforcement Learning for Personalized Dialog Generation: An Empirical Study43
- Hierarchical human-like strategy for aspect-level sentiment classification with sentiment linguistic knowledge and reinforcement learning43
This work proposes a novel Hierarchical Human-like strategy for Aspect-level Sentiment classification (HHAS), which contains three major components, a sentiment-aware mutual attention module, an aspect-specific knowledge distillation module, and a reinforcement learning based re-reading module which are consistent with the stages of the human reading cognitive process.
- Generative adversarial fusion network for class imbalance credit scoring43
An imbalanced generative adversarial fusion network (IGAFN) is proposed to cope with the class imbalance credit scoring based on multi-source heterogeneous credit data and exhibits significantly better performance than the compared methods on two real-life datasets.
- CBN: Constructing a clinical Bayesian network based on data from the electronic medical record43
This study demonstrates that the direct and automated construction of a high-quality health topology and ontology from medical records is feasible and is reproducible.
- 42
- Summarize before Aggregate: A Global-to-local Heterogeneous Graph Inference Network for Conversational Emotion Recognition42
A two-stage Summarization and Aggregation Graph Inference Network (SumAggGIN) is proposed, which seamlessly integrates inference for topic-related emotional phrases and local dependency reasoning over neighbouring utterances in a global-to-local fashion and outperforms the state-of-the-art approaches.
- Contrastive Learning with Hard Negative Entities for Entity Set Expansion41
The ProbExpan is proposed, a novel probabilistic ESE framework utilizing the entity representation obtained by the aforementioned language model to expand entities and outperforms previous state-of-the-art methods.
- 41
- Plausibility-promoting generative adversarial network for abstractive text summarization with multi-task constraint40
A plausibility-promoting Generative Adversarial Network for Abstractive Text Summarization with Multi-Task constraint (PGAN-ATSMT), which shows promising performance for generating informative, grammatically correct, and novel summaries.
- Aspect and Sentiment Aware Abstractive Review Summarization40
An interactive attention mechanism to interactively learns the representations of context words, sentiment words and aspect words within the reviews, acted as an encoder to generate aspect/sentiment-aware abstractive review summaries via an attention fusion network.
- On the Convergence of Zeroth-Order Federated Tuning for Large Language Models39
This study is the first to examine the theoretical underpinnings of FedMeZO in the context of LLMs, tackling key questions regarding the influence of large parameter spaces on optimization behavior, the establishment of convergence properties, and the identification of critical parameters for convergence to inform personalized federated strategies.
- Imagine, Reason and Write: Visual Storytelling with Graph Knowledge and Relational Reasoning39
This paper proposes a novel imagine-reason-write generation framework (IRW) for visual storytelling, inspired by the logic of humans when they write the story, which significantly outperforms the state-of-the-art methods across multiple evaluation metrics.
- X-Eval: Generalizable Multi-aspect Text Evaluation via Augmented Instruction Tuning with Auxiliary Evaluation Aspects38
X-Eval is introduced, a two-stage instruction tuning framework to evaluate text in both seen and unseen aspects customized by end users that enables even a lightweight language model to achieve a comparable if not higher correlation with human judgments compared to the state-of-the-art NLG evaluators like GPT-4.
- Cooperative Denoising for Distantly Supervised Relation Extraction38
CORD, a novelCOopeRativeDenoising framework, which consists two base networks leveraging text corpus and knowledge graph respectively, and a cooperative module involving their mutual learning by the adaptive bi-directional knowledge distillation and dynamic ensemble with noisy-varying instances achieves substantial improvement over state-of-the-art methods.
- Understanding User Behavior in Sina Weibo Online Social Network: A Community Approach37
An information-theoretical measure based on the concept of entropy is introduced to analyze the temporal tweeting behaviors of users and applies a clustering algorithm to divide users into different groups based on their tweeting behaviors, which can improve the design of plenty of applications, such as recommendation systems.
- Answer-driven Deep Question Generation based on Reinforcement Learning36
An Answer-driven Deep Question Generation model based on the encoder-decoder framework with a semantic-rich fusion attention mechanism is designed to support the decoding process, which integrates the answer with the document representations to promote the proper handling of answer information during generation.
- Interactive Dual Generative Adversarial Networks for Image Captioning35
This paper proposes an Interactive Dual Generative Adversarial Network (IDGAN) for image captioning, which mutually combines the retrieval-based and generation-based methods to learn a better image Captioning ensemble.
- Knowledge as A Bridge: Improving Cross-domain Answer Selection with External Knowledge35
This paper proposes Knowledge-aware Attentive Network (KAN), a transfer learning framework for cross-domain answer selection, which uses the knowledge base as a bridge to enable knowledge transfer from the source domain to the target domains.
- Nonfactoid Question Answering as Query-Focused Summarization With Graph-Enhanced Multihop Inference34
A novel query-focused summarization method, namely a graph-enhanced multihop query- focused summarizer (GMQS), to tackle the nonfactoid QA problem and possesses the capability of performing explainable multihip reasoning.
- 33
- The Art of SOCRATIC QUESTIONING: Recursive Thinking with Large Language Models32
The qualitative analysis clearly shows that the intermediate reasoning steps elicited by SOCRATIC QUESTIONING are similar to humans' recursively thinking process of complex reasoning problems.
- Contextual-boosted deep neural collaborative filtering model for interpretable recommendation32
A Contextual-boosted Deep Neural Collaborative filtering model for item recommendation, which simultaneously exploits both item introductions and user ratings to alleviate the cold-start problem and provide interpretable item recommendation.
- Cross-domain aspect/sentiment-aware abstractive review summarization by combining topic modeling and deep reinforcement learning32
The novel model Abstractive review Summarization with Topic modeling and Reinforcement deep learning (ASTR) leverages the benefits of the supervised deep neural networks, reinforcement learning, and unsupervised probabilistic generative model to strengthen the aspect/sentiment-aware review representation learning.
- Automatic Context Pattern Generation for Entity Set Expansion31
The GAPA is proposed, a novel ESE framework that leverages the aforementioned GenerAted PAtterns to expand target entities and devise a context pattern generation module that utilizes autoregressive language models (e.g., GPT-2) to automatically generate high-quality context patterns for entities.
- 31
- KGDDS: A System for Drug-Drug Similarity Measure in Therapeutic Substitution based on Knowledge Graph Curation31
A system KGDDS for node-link-based bio-medical Knowledge Graph curation and visualization, aiding Drug-Drug Similarity measure, and a user interaction model to allow the user to better understand the drug properties from a drug similarity perspective and gain insights that are not easily observable in individual drugs.
- Global trends and risk factors of laryngeal cancer: a systematic analysis for the Global Burden of Disease Study (1990–2021)30
From 1990 to 2021, global LC incidence, deaths, and DALYs decreased, although regional disparities persisted, and death rates remain high in economically disadvantaged regions, especially among the elderly and females with occupational risk.
- 30
- 29
- Contextualized Knowledge-aware Attentive Neural Network: Enhancing Answer Selection with Knowledge28
A context-knowledge interaction learning framework, Knowledge-aware Neural Network, which learns the QA sentence representations by considering a tight interaction with the external knowledge from KG and the textual information and develops two kinds of knowledge-aware attention mechanism to summarize both the context-based and knowledge-based interactions between questions and answers.
- Exploring Human-Like Reading Strategy for Abstractive Text Summarization28
This work proposes a novel Hybrid learning model for Abstractive Text Summarization (HATS), which consists of three major components, a knowledge-based attention network, a multitask encoder-decoder network, and a generative adversarial network, which are consistent with the different stages of the human-like reading strategy.
- 27
- De-identification of Clinical Text via Bi-LSTM-CRF with Neural Language Models.27
This paper introduces Bi-LSTM-CRF with neural language models for de- identification of clinical text, and evaluates it on the de-identification datasets of the i2b2 2014 and the CEGS N- GRID 2016 challenges.
- Amalgamating Knowledge from Two Teachers for Task-oriented Dialogue System with Adversarial Training25
This paper proposes a “Two-Teacher One-Student” learning framework (TTOS) for task-oriented dialogue, with the goal of retrieving accurate KB entities and generating human-like responses simultaneously.
- Enhancing ontology-driven diagnostic reasoning with a symptom-dependency-aware Naïve Bayes classifier24
The experimental results demonstrate that the proposed method can effectively identify accurate medical knowledge probability information from EMR data, thereby demonstrating the advantage of combining an ontology and a symptom-dependency-aware naïve Bayes classifier.
- KMR: knowledge-oriented medicine representation learning for drug–drug interaction and similarity computation24
Experimental results indicate that KMR can identify DDS with an accuracy rate of 88.7% by facilitating drug knowledge, outperforming existing state-of-the-art drug similarity measures.
- 23
- Cross-modal Image-Text Retrieval with Multitask Learning23
Two regularization terms (variance and consistency constraints) are introduced to the cross-modal embeddings such that the learned common information has large variance and is modality invariant and to enable large-scale cross- modal similarity search, a flexible binary transform network is designed to convert the text and imageembeddings into binary codes.
- Cross-domain Aspect/Sentiment-aware Abstractive Review Summarization23
The proposed model CASAS (neural attentive model for Cross-domain Aspect/Sentiment-aware Abstractive review Summarization) leverages domain classification task, working on datasets of both source and target domains, to recognize the domain information of texts and transfer knowledge from source domains to target domains.
- From Training-Free to Adaptive: Empirical Insights into MLLMs’ Understanding of Detection Information22
A systematic study comparing three strategies: training-free infusion, full retraining, and lightweight fine-tuning reveals that a simple fine-tuning strategy is remarkably effective, outperforming its training-free counterpart by a significant margin.
- Hierarchical fusion of common sense knowledge and classifier decisions for answer selection in community question answering22
A novel Knowledge-enhanced Hierarchical Attention mechanism for Answer Selection (KHAAS), which fully exploits the common sense knowledge from knowledge bases and input textual information to learn the fact-aware intra-document features within questions and answers by fusing the knowledge from both the question/answer and KB.
- Knowledge-enhanced Hierarchical Attention for Community Question Answering with Multi-task and Adaptive Learning22
A Knowledge-enhanced Hierarchical Attention for community question answering with Multi-task learning and Adaptive learning (KHAMA) and a hierarchical attention network to fully fuse knowledge from input documents and knowledge base by exploiting the semantic compositionality of the input sequences.
- Improving Medical Short Text Classification with Semantic Expansion Using Word-Cluster Embedding21
This work proposes to add word-cluster embedding to deep neural network for improving short text classification and demonstrates that the proposed method outperforms state-of-the-art baselines in short sentence classification on both medical domain and general domain.
- Img-Diff: Contrastive Data Synthesis for Multimodal Large Language Models19
A novel data synthesis method, leveraging insights from contrastive learning and image difference captioning to enhance fine-grained image recognition in MLLMs, yielding comprehensive improvements across numerous image difference and Visual Question Answering tasks.
- Multimodal single-cell analyses reveal molecular markers of neuronal senescence in human drug-resistant epilepsy18
By integrating whole-cell patch clamp recording and single-cell RNA sequencing approaches, a transcriptionally distinct subset of cortical pyramidal neurons is identified that highly expressed genes CDKN1A (P21), CCL2 and NFKBIA that associated with mTOR pathway, inflammatory response and cellular senescence are identified.
- Dynamic Demonstration Retrieval and Cognitive Understanding for Emotional Support Conversation18
This paper introduces Dynamic Demonstration Retrieval and Cognitive-Aspect Situation Understanding (D2RCU), a novel approach that synergizes these elements to improve the quality of support provided in ESCs and proposes a cognitive understanding module that utilizes four cognitive relationships from the ATOMIC knowledge source to deepen situational awareness of help-seekers' mental states.
- Bridging Hierarchical and Sequential Context Modeling for Question-driven Extractive Answer Summarization18
This paper designs a hierarchical compare-aggregate method to integrate the interaction between QA pairs in both word-level and sentence-level into the final question and answer representations and conducts the question-aware sequential extractor to produce a summary for the lengthy answer.
- 18
- 18
- Path-based reasoning with constrained type attention for knowledge graph completion17
This work develops a novel path-based reasoning with constrained type attention model, which tries to identify entity types by leveraging relation type constraints in the corresponding triples of knowledge graph completion.
- Attentive User-Engaged Adversarial Neural Network for Community Question Answering16
An Attentive User-engaged Adversarial Neural Network (AUANN) is presented, which interactively learns the context information of questions and answers, and enhances user engagement with the CQA task.
- Attention-Aware Path-Based Relation Extraction for Medical Knowledge Graph16
Compared with the state-of-the-art relation extraction methods using the structure of knowledge graph, the proposed method significantly improves the accuracy of extracted relation facts and achieves the best performance.
- 15
- MedTruth15
A new truth discovery method is proposed, MedTruth, for medical knowledge condition discovery, which incorporates prior source quality information into the source reliability estimation procedure, and also utilizes the knowledge triple information for trustworthy information computation.
- 14
- Relabel the Noise: Joint Extraction of Entities and Relations via Cooperative Multiagents14
This paper considers the problem of shifted label distribution, which is caused by the inconsistency between the noisy-labeled training set subject to external knowledge graph and the human-annotated test set, and exacerbated by the pipelined entity-then-relation extraction manner with noise propagation.
- Constructing Ontology-Based Cancer Treatment Decision Support System with Case-Based Reasoning14
The proposed DSS uses Case-Based Reasoning (CBR) to consider disease manifestations and provides physicians with treatment solutions from similar previous cases for reference and supports natural language processing (NLP) queries.
- 14
- 14
- Optimal classification for the diagnosis of duchenne muscular dystrophy images using support vector machines14
The T1W images in SVM-based classifier at level 2 decomposition showed the highest performance of all, demonstrating that it was the optimal classification for the diagnosis of DMD.
- Discovering author interest evolution in order-sensitive and Semantic-aware topic modeling13
An Ordering-sensitive and Semantic-aware Dynamic Author Topic Model (OSDATM) is proposed, which monitors the evolution of author interest in time-stamped documents and uses the discovered author interest information to discover better topics.
- MARES: multitask learning algorithm for Web-scale real-time event summarization13
A novel algorithm for Web-scale real-time summarization of massive document streams, which leverages the benefits of supervised deep neural networks as well as a reinforcement learning algorithm to strengthen the representation learning of documents.
- 12
- Learning to Augment for Data-scarce Domain BERT Knowledge Distillation12
This work proposes a method to learn to augment data for BERT Knowledge Distillation in target domains with scarce labeled data, by learning a cross-domain manipulation scheme that automatically augments the target domain with the help of resource-rich source domains.
- Practical fine-grained learning based anomaly classification for ECG image12
A novel neural framework is designed that can be directly applied to massive ECG images determining various types of cardiology abnormalities and can result in significant savings in annotation time and cost by eliminating the need for part annotations.
- 11
- Investigating the spontaneous brain activities of patients with subjective cognitive decline and mild cognitive impairment: an amplitude of low-frequency fluctuation functional magnetic resonance imaging study11
Spontaneous brain activity is a stable imaging biomarker of cognitive impairment and ALFF changes of the prefrontal, occipital, left angular, and temporal gyrus were sensitive to identifying cognitive decline, and the scores of the Auditory-Verbal Learning Test and MoCA could predict the abnormal intrinsic activities.
- Multi-source Domain Adaptation for Sentiment Classification with Granger Causal Inference11
A multi-source domain adaptation method with a Granger-causal objective (MDA-GC) for cross-domain sentiment classification and results demonstrate that the proposed MDA- GC model significantly outperforms the compared methods.
- 11
- Knowledge-aware Textual Entailment with Graph Attention Network11
A Knowledge-Context Interactive Textual Entailment Network (KCI-TEN) that learns graph level sentence representations by harnessing external knowledge graph with graph attention network and a text-graph interaction mechanism for neural based entailment matching learning that endows the redundancy and noise with less importance and put emphasis on the informative representations.
- Using social media to assess the consumer nutrition environment: comparing Yelp reviews with a direct observation audit instrument for grocery stores11
Yelp holds promise as a cost-effective means to gather information on the overall cost, quality and experience of food stores, which may be relevant for nutrition outcomes.
- Improving Semi-Supervised Text Classification with Dual Meta-Learning10
A dual meta-learning (DML) technique for semi-supervised text classification is introduced, which improves the teacher and student classifiers simultaneously in an iterative manner and devise a meta pseudo supervision method to improve the teacher classifier.
- A Unified Framework for Contextual and Factoid Question Generation10
A unified framework for the tasks of CQG and FQG is proposed, where two types of task-sharing modules are developed to learn shared contextual and structural knowledge, and the task format is unified with a pseudo passage reformulation strategy.
- 10
- Browsing Like Human: A Multimodal Web Agent with Experiential Fast-and-Slow Thinking9
This work proposes a novel multimodal web agent framework called WebExperT, which is designed to emulate the human planning process of "thinking fast and slow" to effectively decompose complex user instructions.
- STARFlow-V: End-to-End Video Generative Modeling with Normalizing Flows9
These results present the first evidence, to the authors' knowledge, that NFs are capable of high-quality autoregressive video generation, establishing them as a promising research direction for building world models.
- 9
- Integrating User History into Heterogeneous Graph for Dialogue Act Recognition9
A Heterogeneous User History (HUH) graph convolution network, which utilizes the user’s historical answers grouped by DA labels as additional clues to recognize the DA label of utterances, and outperforms the state-of-the-art methods.
- 9
- Many-to-many Image Generation with Auto-regressive Diffusion Models8
A domain-general framework for many-to-many image generation, capable of producing interrelated image series from a given set of images, offering a scalable solution that obviates the need for task-specific solutions across different multi-image scenarios.
- HRKD: Hierarchical Relational Knowledge Distillation for Cross-domain Language Model Compression8
To enhance the model capability and transferability, the idea of meta-learning is leveraged and set up domain-relational graphs to capture the relational information across different domains, and to dynamically select the most representative prototypes for each domain, a hierarchical compare-aggregate mechanism to capture hierarchical relationships.
- Be Relevant, Non-Redundant, and Timely: Deep Reinforcement Learning for Real-Time Event Summarization8
A Deep Reinforcement learning framework for real-time Event Summarization (DRES), which shows promising performance for resolving all three challenges (i.e., relevance, non-redundancy, timeliness) in a unified framework.
- 7
- 7
- Wasserstein Selective Transfer Learning for Cross-domain Text Mining7
The proposed Wasserstein Selective Transfer Learning (WSTL) method considers a reinforced selector to select helpful data for transfer learning and adopts an evaluation metric based on the performance of the TL module as delayed reward and aWasserstein-based metric as immediate rewards to guide the reinforced selector learning.
- Gastroenterology Ontology Construction Using Synonym Identification and Relation Extraction7
The proposed synonym identification and relation extraction method achieves the best performance compared with state-of-the-art methods and also builds a more complete ontology compared with existing gastroenterology disease ontologies.
- 7
- 6
- Correct Like Humans: Progressive Learning Framework for Chinese Text Error Correction6
A novel model-agnostic progressive multi-task learning framework for CTEC is proposed, named ProTEC, which guides a CTEC model to learn the task from easy to difficult, and divides CTEC task into three sub-tasks from easy to difficult: Error Detection, Error Type Identification, and Correction Result Generation.
- 6
- ANGraph: attribute-interactive neighborhood-aggregative graph representation learning6
An Attribute-interactive Neighborhood-aggregative Graph learning scheme (ANGraph), which better preserves structure proximity and attribute affinity by leveraging attribute interaction and smoothness measures to incorporate vertices similar/close to each other in the original space is proposed.
- Collaborating Heterogeneous Natural Language Processing Tasks via Federated Learning6
This study further broaden the application scope of FL in NLP by proposing an Assign-Then-Contrast (denoted as ATC) framework, which enables clients with heterogeneous NLP tasks to construct an FL course and learn useful knowledge from each other.
- Integrating Linguistic Knowledge to Sentence Paraphrase Generation6
Knowledge-Enhanced Paraphrase Network (KEPN) is proposed, a transformer-based framework that can leverage external linguistic knowledge to facilitate paraphrase generation and significantly outperforms the state-of-the-art approaches in terms of both automatic and human evaluation.
- A Multi-Task Learning Approach for Answer Selection: A Study and a Chinese Law Dataset6
The experimental results show that the model MTAS consistently outperforms the compared methods.
- MedSim: A Novel Semantic Similarity Measure in Bio-medical Knowledge Graphs6
A novel semantic SIMilarity method based on public well-established bio-MEDical knowledge graphs and large-scale corpus to study the therapeutic substitution of antibiotics and some promising applications of MedSim in drug substitution and drug abuse prevention are presented.
- EPICE an emotion fuzzy vectorial space for time modeling in medical decision6
A fuzzy vectorial space is proposed to model the emotion felt by the patient during his care course and the relationship with the caregivers to avoid fuzzification and denazification steps of the rule-based approaches and allows time modeling.
- An enhanced computational feature selection method for medical synonym identification via bilingualism and multi-corpus training6
This paper selected 13 features including Chinese and English features and studied the synonym identification results of each feature alone and different combinations of the features to present an optimal combination of features for Chinese medical synonyms identification.
- 6
- 6
- LTCR: Long-Text Chinese Rumor detection dataset5
A Long-Text Chinese Rumor detection dataset named LTCR is proposed, which provides a valuable resource for accurately detecting misinformation, especially in the context of complex fake news related to COVID-19.
- InScope: A new real-world 3D infrastructure-side collaborative perception dataset for open traffic scenarios5
This paper introduces a new 3D infrastructure-side collaborative perception dataset, abbreviated as InScope, which is the first dataset dedicated to addressing occlusion challenges by strategically deploying multiple-position Light Detection and Ranging systems on the infrastructure side.
- Mitigating Catastrophic Forgetting in Multi-domain Chinese Spelling Correction by Multi-stage Knowledge Transfer Framework5
A novel model-agnostic Multi-stage Knowledge Transfer (MKT) framework is proposed, which utilizes a continuously evolving teacher model for knowledge transfer in each domain, rather than focusing solely on new domain knowledge.
- Hierarchical Type Enhanced Negative Sampling for Knowledge Graph Embedding5
Hierarchical Type Enhanced Negative Sampling (HTENS) is proposed which leverages hierarchical entity type information and entity-relation cooccurrence information to optimize the sampling probability distribution of negative samples.
- 5
- Abstractive Text Summarization with Hierarchical Multi-scale Abstraction Modeling and Dynamic Memory5
A novel abstractive text summarization method with hierarchical multi-scale abstraction modeling and dynamic memory (called MADY) to capture the temporal dependencies of the document from multiple hierarchical levels of abstraction, which mimics the process of how human beings comprehend an article.
- Sunlight Pollution Analysis of Glass Curtain Wall in 3D City5
The method ensures instantaneity of urban sunlight pollution acquisition and computation, to a certain extent covers shortages of current researches, and is capable of reference evaluation in environment protection, accurate warnings and predictions of glare intensity for pedestrians and drivers.
- Answer-enhanced Path-aware Relation Detection over Knowledge Base5
A Knowledge-driven Relation Detection network (KRD) is proposed to interactively learn answer-enhanced question representations and path-aware relation representations for relation detection and experimental results demonstrate that KRD outperforms the state-of-the-art methods.
- EAPB: entropy-aware path-based metric for ontology quality5
An Entropy-Aware Path-Based (EAPB) metric for ontology quality is proposed by considering the path information between different vertices and textual information included in the path to calculate the connectivity path of the whole network and dynamic weights between different nodes.
- Study on Dynamic Reliability Analysis of Materiel System Based on Adaptive Agent Simulation5
A system structure model based on Agent, a mission-oriented reliability impact model and a dynamic reliability model are constructed which help to simulate and analyze dynamic reliability action oriented on mission environment, effectively settle reliability engineering problems, with the application and amendment of the models.
- 4
- ELBA: Learning by Asking for Embodied Visual Navigation and Task Completion4
Experimental results show that the proposed Embodied learning-By-Asking model achieves improved task performance compared to baseline models without question-answering capabilities.
- 4
- 4
- 4
- 4
- Modeling path information for knowledge graph completion4
An Attribute-embodied neural Relation Path Prediction model is proposed to predict missing relations between entities in a KG to enrich the representation learning and aid in learning more valuable information from noise paths for relation prediction.
- An in-depth study of similarity predicate committee4
This paper proposes a novel approach to combine different similarity predicates together to form a committee so that one does not need to worry about choosing which of them to use, and can obtain a better result than any individual similarity predicate.
- A Knowledge Graph Based Solution for Entity Discovery and Linking in Open-Domain Questions4
This work proposed a knowledge graph based solution for QEDL and developed a system consisting of Question Entity Discovery (QED) module and Entity Linking (EL) module that could obtain 64.44% F1 score of QED and 64.86% accuracy of EL, which ranks the 2nd place and indicates its practical use forQEDL problem.
- 4
- When Truth Discovery Meets Medical Knowledge Graph: Estimating Trustworthiness Degree for Medical Knowledge Condition.4
A new truth discovery method is proposed to explore medical-related texts and infer trustworthiness degrees of knowledge triples associating with different conditions and demonstrates the effectiveness of the proposedtruth discovery method.
- 3
- 3
- Learning Purified Feature Representations from Task-irrelevant Labels3
This work purify feature representations by using the expression of task-irrelevant information, thus facilitating the learning process of classification, and proves that Purified Learning is model-agnostic and can be combined with any existing deep neural networks with ease to achieve better performance.
- Forecasting Stock Index Using a Volume-Aware Positional Attention-Based Recurrent Neural Network3
A volume-aware positional attention-based recurrent neural network (VPA-RNN) for stock performance prediction, and the experimental results show that the proposed VPA-RNN can significantly outperform several existing highly competitive methods.
- 3
- 3
- Learning Document Embeddings with Crossword Prediction3
A Document Embedding Network (DEN) to learn document embeddings in an unsupervised manner using the encoder-decoder architecture as its backbone, which tries to reconstruct the input document from an encoded document embedding.
- 3
- How to Perfect the Interpretation and Translation in Engineering Projects3
Through analyzing the engineering properties from the angle of Translation, this article shows that, due to its distinctiveness, the difficulty of proper Interpretation and Translation in Engineering Projects mainly rests with accurately understanding the meaning of source language and correctly using specialized vocabulary to effectively translate the original meaning to audience.
- Vibration stimulation enhances robustness in teleoperation robot system with EEG and eye-tracking hybrid control2
The vibration stimulation technique proposed in this study can effectively improve the training efficiency and online decoding rate of MI, helping users enhance their control efficiency while focusing on control tasks.
- Identification of potential molecular targets for thyroid cancer using multiple-omics analysis and machine learning2
FN1 and CRABP1 were identified as potential molecular targets in thyroid cancer, which influence tumor progression, immune response, and patient prognosis, offering new insights for early diagnosis and personalized treatment strategies.
- 2
- 2
- 2
- Cross-perspective Graph Contrastive Learning2
Inspired by the network homophily, it is argued that there is a deep correlation information between the network structure perspective and the node attributes perspective and the potential information between them is exploited.
- 2
- 2
- Ontology Evaluation with Path-based Text-aware Entropy Computation2
This study proposes a Path-based Text-aware Entropy Computation method, PTEC, by considering the path information between different vertices and the textual information within the path to calculate the connectivity path of the whole network and the different weights between various nodes.
- 2
- Automated Evaluation of Text and Discourse with Range2
Analysis of vocabulary use in relation to reading comprehension in Test for English Majors-Band 4 and Band 8 from 2000 to 2014 finds pronouns, conjunctions and propositions are used more frequently than other word categories, such as modal auxiliaries.
- 2
- 1
- 1
- 1
- FIT-graph: A multi-grained evolutionary graph based framework for disease diagnosis1
FIT-Graph efficiently organizes the extracted information from various granularities and time stages, maximizing the retention of valuable information for disease inference and ensuring the comprehensiveness and validity of the final disease inference.
- 1
- Counterfactual Debiasing for Generating Factually Consistent Text Summaries1
This paper constructs causal graphs for abstractive text summarization and identifies the intrinsic causes of the factual inconsistency, i.e., the language bias and irrelevancy bias, and proposes a debiasing framework, named CoFactSum, to alleviate the causal effects of these biases by counterfactual estimation.
- 1
- SiD-WaveFlow: A Low-Resource Vocoder Independent of Prior Knowledge1
Experimental results have corroborated that SiD-WaveFlow can generate speech with better quality compared with its counterparts, and SiD-WaveFlow is a lightweight model and can generate speech on edge devices with a much faster inference speed.
- 1
- An Adaptive Embedding Framework for Heterogeneous Information Networks1
A novel adaptive embedding framework is proposed, which integrates these two kinds of methods to preserve both topological information and relational information by incorporating an assistant knowledge graph embedding model.
- 1
- Feature-Aware Attentive Convolutional Neural Network for Sequence Processing1
A feature-aware Sequence Attentive Convolutional Neural Network (SeqANN) that interactively learns the sequence representations by harnessing the sequence global and local information as well as the expressive domain knowledge is proposed.
- Semantic Similarity Measures to Disambiguate Terms in Medical Text1
An approach MedSim is presented for semantic similarity measures to identify synonym terms in medical text with effectiveness and accuracy well-balanced and experimental results demonstrate that the proposed method has robust superiority over competitors for synonym identification.
- Approach for Semi-automatic Construction of Anti-infective Drug Ontology Based on Entity Linking1
In this study, an NLP approach is adopted to generate candidate entities and six-word vector features and word-level features are selected to perform the entity linking.
- An event summarizing algorithm based on the timeline relevance model in Sina Weibo1
A timeline relevance model was initially established to estimate the popular period of the event, and a summarizing algorithm was designed to mine and summarize the related events to help users to quickly identify and grasp the event that is related to the topic in a limited time frame.
- The procedure and challenges for pediatric reference interval establishment in clinical laboratory1
The procedure and key technologies of direct method and indirect method are stated based on the characteristics of children population and pediatric, by which to define, transfer and validate pediatric reference intervals.
- 1
- Study on System Reliability Allocation Based on Adaptive Agent Simulation1
The method applies to either basic reliability allocation or mission reliability allocation, considers the change of system environment to the effect on mission, and determines system reliability allocation conception oriented on mission with relative optimization and better application.
- 1
- 1
- 1
- Multi-scale Representation of Spatial Information on Voronoi Images1
According to principles of Voronoi and Delaunary of computational geometry, the selection of point elements, the intersection of line elements, and the converging and predigesting of area elements were studied to make them more favorable to the human visual and perceptive system, and consequently getting the spatial data more propitious to analyze, process and visualize.
- Vibrotactile stimulation for upper-limb neurorehabilitation and sensorimotor training: a systematic review of clinical and mechanistic evidence–
Current evidence suggests that VTS may support upper-limb functional recovery after stroke, however, confidence in the evidence remains limited by small samples, heterogeneous protocols, and the frequent use of multimodal interventions that do not isolate the contribution of vibration.
- –
- –
- –
- –
- FLEX-CLIP: Feature-Level GEneration Network Enhanced CLIP for X-shot Cross-modal Retrieval–
FLEX-CLIP is proposed, a novel Feature-level Generation Network Enhanced CLIP that combines a composite multimodal VAE-GAN network to capture real feature distribution patterns and generate pseudo samples based on CLIP features, addressing data imbalance.
- –
- –
- –
- –
- –
- –
- Multi-Task Learning for Symptom Name Recognition and Severity Assessment in Electronic Medical Records (Preprint)–
A novel cluster-based knowledge-aware learning scheme to reduce semantic ambiguity for name recognition and enrich sentence representation learning for severity assessment is proposed, which can provide symptom knowledge and implications for clinicians and patients as a reference and has remarkable applicability and generality.
- Study on the Construction Frame of Materiel Reliability Simulation Data Support Environment Based on Symposium of Metasynthetic Engineering–
The paper studies the theory of hall for symposium of metsynthetic Engineering the and constructs the frame of data support environment for materiel reliability simulation, to provide data support and verify the results of simulation.
- Research on Data Analysis of Material System Based on Agent Simulation–
In the paper, it sets up a fine mission reliability simulation model of materiel system, and analyzes from data collection and process, to lay the good foundation for analyzing dynamically mission reliability and making a perfect mission reliability Simulation concept of materel system.
- RANGE Application in Examination-oriented Education: Focusing on TEM-8 Translations–
This thesis aims to help the students English learners especially English majors further lay the foundation and enlarge their vocabulary to improve translation grades in TEM-8 by collecting and analyzing original tests and corresponding drills via RANGE, and figure out the types of words that most commonly used and some skills in word usage.
- –
- The Technique to Design Sharing Maintenance Resources among Maintenance Modules–
To promote modularized maintenance force building, the reasons to share resources among maintenance modules and the standards to design share resources are analyzed, based on the kinds of classification result of maintenance modules.
- –
- Assessment of Multiple Working Condition System Reliability with Agent-Based Simulation Method–
A multi-working condition simulation agent was designed and used to simulate the random transferring process of working conditions of system, and it cooperated with system simulation agents and unit simulation agents to realize system mission reliability(MR)simulation.
- Application of Corpus Analysis Tool in Investigating Listening Proficiency: Under the Perspective of TEM-8 Testing by RANGE–
This study used qualitative and quantitative approaches to analyze the listening exam items in Test for English Majors-Band 8 (TEM-8) from 1996 to 2010, finding the frequency of the tokens and types in the entire listening tests is found.
- –
- –
- –
- Level of Repair Analysis Oriented on Integrated Materiel Support Based on BP Neural Network–
The paper discusses how to go on LORA in the context of integrated materiel support based on BP neural network, better to settle the problem of LORN, and puts the analysis technology into practice in deed.
- –
- Study on Framework of Environment Adaptive Agent Based Materiel System Reliability Simulation Modeling–
The framework provides a current technology concept for adaptive Agent Based Materiel System Reliability Simulation, with better universality and maneuverability.
- –
- –
- –
- –
- –
- –
- –
- Automatic Data Recovery Technology with System Parameter Detection–
This work introduced an automatic data recovery technology with system parameter detection, which granted a higher reliability and usability to current data recovery.
- Research on Tolerance Design of Weapon Simulation System–
The reliability of the system can be improved by combination of hardware and software in the tolerance design of the weapon simulation system on the basis of the embedded computer.
- Distributed Generation of Shared RSA Keys in Mobile Ad Hoc Networks–
This paper proposes the solution of distributed generation of shared CA keys based on threshold RSA cryptosystems, with which the security and robustness of system is enhanced.
- san qi zao zuo dui man xing shen yan huan zhe wai zhou xue t lin ba xi bao ya qun ji xie tong ci ji fen zi cd2cd58 biao da diao jie de yan jiu–
Familiarity, ease of access, trust, and awareness of benefits and risks are important.
- –
- Compiling a synchronous programming language into field programmable gate arrays–
This thesis shows how to compile a program expressed by a novel hardware description language, the State Machine Algol-Like Language (SMALL), into Field- Programmable Gate Arrays (FPGAs).
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-14. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.