Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- Rong JinSuggested from co-authorship
- Jamie CallanSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works320 from public data
- Knowing What, How and Why: A Near Complete Solution for Aspect-Based Sentiment Analysis531
This paper introduces a new subtask under ABSA, named aspect sentiment triplet extraction (ASTE), and proposes a two-stage framework to address this task, which outperforms a few strong baselines adapted from state-of-the-art related methods.
- KnowPrompt: Knowledge-aware Prompt-tuning with Synergistic Optimization for Relation Extraction518
A Knowledge-aware Prompt-tuning approach with synergistic optimization (KnowPrompt) that injects latent knowledge contained in relation labels into prompt construction with learnable virtual type words and answer words.
- An automatic weighting scheme for collaborative filtering371
An optimization algorithm to automatically compute the weights for different items based on their ratings from training users will create a clustered distribution for user vectors in the item space by bringing users of similar interests closer and separating users of different interests more distant.
- Flexible mixture model for collaborative filtering356
FMM extends existing partitioning/clustering algorithms for collaborative filtering by clustering both users and items together simultaneously without assuming that each user and item should only belong to a single cluster.
- A statistical model for scientific readability293
Experiments show that this new method of using statistical models to estimate readability has a better performance than the widely used Flesch-Kincaid readability formula.
- Relevant document distribution estimation method for resource selection284
It is shown that the CORI algorithm does not do well in environments with a mix of "small" and "very large" databases, and a new resource selection algorithm is proposed that uses information about database sizes as well as database contents.
- CBLUE: A Chinese Biomedical Language Understanding Evaluation Benchmark262
The first Chinese Biomedical Language Understanding Evaluation Evaluation (CBLUE) benchmark is presented: a collection of natural language understanding tasks including named entity recognition, information extraction, clinical diagnosis normalization, single-sentence/sentence-pair classification, and an associated online platform for model evaluation, comparison, and analysis.
- Hybrid Transformer with Multi-level Fusion for Multimodal Knowledge Graph Completion260
This paper proposes a hybrid transformer architecture with unified input-output for diverse multimodal knowledge graph completion tasks, and proposes multi-level fusion, which integrates visual and text representation via coarse-grained prefix-guided interaction and fine- grained correlation-aware fusion modules.
- On the Effectiveness of Adapter-based Tuning for Pretrained Language Model Adaptation260
It is demonstrated that 1) adapter-based tuning outperforms fine-tuning on low-resource and cross-lingual tasks; 2) it is more robust to overfitting and less sensitive to changes in learning rates.
- Expertise Retrieval260
This survey highlights advances in models and algorithms relevant to expertise retrieval as an emerging subdiscipline in information retrieval and draws connections among methods proposed in the literature and summarizes them in five groups of basic approaches.
- Document-level Relation Extraction as Semantic Segmentation236
A Document U-shaped Network for document-level relation extraction is proposed, which leverages an encoder module to capture the context information of entities and a U- shaped segmentation module over the image-style feature map to capture global interdependency among triples.
- Composite hashing with multiple information sources225
The focus of the new research problem is to design an algorithm for incorporating the features from different information sources into the binary hashing codes efficiently and effectively, and to propose an algorithm CHMIS-AW (CHMIS with Adjusted Weights) for learning the codes.
- HERCULE: attack story reconstruction via community discovery on correlated log graph190
HerCULE is an automated multi-stage log-based intrusion analysis system inspired by graph analytics research in social network analysis that can reconstruct attack behaviors from a spectrum of cyber attacks that involve multiple stages with high accuracy and low false positive rates.
- GALAXY: A Generative Pre-trained Model for Task-Oriented Dialog with Semi-supervised Learning and Explicit Policy Injection187
GALAXY is a novel pre-trained dialog model that explicitly learns dialog policy from limited labeled dialogs and large-scale unlabeled dialog corpora via semi-supervised learning and has a stronger few-shot ability than existing models under various low-resource settings.
- 171
- 167
- Rumor Detection by Exploiting User Credibility Information, Attention and Multi-task Learning167
A new multi-task learning approach for rumor detection and stance classification tasks with a shared layer and two task specific layers that outperforms the state-of-the-art rumor detection approaches.
- A language modeling framework for resource selection and results merging165
This paper extends the language modeling approach to integrate resource selection, ad-hoc searching, and merging of results from different text databases into a single probabilistic retrieval model, designed primarily for Intranet environments.
- Perceive Your Users in Depth158
This work proposes to learn universal user representations across multiple tasks for more effective personalization and refers this work as Deep User Perception Network (DUPN) and conducts an extensive set of offline and online experiments.
- Federated Search157
The goal of this work is to provide a comprehensive summary of the previous research on the federated search challenges described above, to provide a comprehensive summary of the previous research on the result merging problem.
- RelationPrompt: Leveraging Prompts to Generate Synthetic Data for Zero-Shot Relation Triplet Extraction155
This work unify language model prompts and structured text approaches to design a structured prompt template for generating synthetic relation samples when conditioning on relation label prompts (RelationPrompt), and designs a novel Triplet Search Decoding method.
- StructuralLM: Structural Pre-training for Form Understanding153
This paper proposes a new pre-training approach, StructuralLM, to jointly leverage cell and layout information from scanned documents, with two new designs to make the most of the interactions of cell and layouts information.
- Effective automatic image annotation via a coherent language model and active learning153
A coherent language model for automatic image annotation is proposed that takes into account the word-to-word correlation by estimating a coherent language models for an image to significantly reduce the required number of annotated image examples.
- A semisupervised learning method to merge search engine results153
This article presents a semisupervised learning solution to the result merging problem and demonstrates that this method is more effective than the well-known CORI result-merging algorithm under a variety of conditions.
- 150
- Mining contrastive opinions on political texts using cross-perspective topic model139
An extensive set of experiments have been conducted to evaluate the proposed unsupervised topic model for contrastive opinion modeling, which simulates the generative process of how opinion words occur in the documents of different collections.
- Relational Learning with Gated and Attentive Neighbor Aggregator for Few-Shot Knowledge Graph Completion136
A novel gated and attentive neighbor aggregator is built for accurately integrating the semantics of a few-shot relation's neighborhood, which helps filtering the noise neighbors even if a KG contains extremely sparse neighborhoods.
- DAGA: Data Augmentation with a Generation Approach for Low-resource Tagging Tasks134
To generate high quality synthetic data for low-resource tagging tasks, a novel augmentation method with language models trained on the linearized labeled sentences is proposed that can consistently outperform the baselines.
- A Neural Multi-digraph Model for Chinese NER with Gazetteers129
This work proposes a novel approach based on graph neural networks with a multi-digraph structure that captures the information that the gazetteers offer that is effective in incorporating rich gazetteser information while resolving ambiguities, outperforming previous approaches.
- 126
- MELM: Data Augmentation with Masked Entity Language Modeling for Low-Resource NER121
This work proposes Masked Entity Language Modeling (MELM) as a novel data augmentation framework for low-resource NER, and explicitly inject NER labels into sentence context, and thus the fine-tuned MELM is able to predict masked entity tokens by explicitly conditioning on their labels.
- Cascade Ranking for Operational E-commerce Search121
A novel cascade ranking model is designed and deployed in an operational e-commerce search application and an extensive set of experiments demonstrate the advantage of the proposed work to address multiple factors of effectiveness, efficiency and user experience in the real-world application.
- Collaborative filtering with decoupled models for preferences and ratings120
This work builds two separate models for collaborative filtering: a preference model capturing which items are favored by the user and a rating model capturing how the user would rate an item given the preference information.
- A study of mixture models for collaborative filtering111
The study shows that graphical models are powerful tools for modeling collaborative filtering, but careful design is necessary to achieve good performance, and proposes three properties that a graphical model is expected to satisfy.
- A Survey on Text-to-SQL Parsing: Concepts, Methods, and Future Directions110
A comprehensive review on deep learning approaches for text-to-SQL parsing, including a systematical overview of pre-trained language models and existing methods, is presented.
- MulDA: A Multilingual Data Augmentation Framework for Low-Resource Cross-Lingual NER110
This paper first proposes a simple but effective labeled sequence translation method to translate source-language training data to target languages and avoids problems such as word order change and entity span determination.
- Aspect Sentiment Classification with both Word-level and Clause-level Attention Networks107
The need for incorporating the importance degrees of both words and clauses inside a sentence is highlighted and a hierarchical network with both word-level and clause-level attentions to aspect sentiment classification is proposed.
- 104
- De-Biased Court’s View Generation with Causality101
A novel Attentional and Counterfactual based Natural Language Generation (AC-NLG) method, consisting of an attentional encoder and a pair of innovative counterfactual decoders, to eliminate the confounding bias in data and generate judgment-discriminative court’s views by incorporating with a synergistic judgment predictive model.
- 98
- Modeling search engine effectiveness for federated search96
A Federated search technique that uses utility maximization to model the retrieval effectiveness of each search engine in a federated search environment is proposed and shown to be effective.
- Matrix co-factorization for recommendation with rich side information and implicit feedback94
This paper proposes matrix factorization techniques to incorporate rich user and item information into recommendation with implicit feedback in online scientific communities which exhibit two characteristics: there exists very rich information about users and items and the users in the scientific communities tend not to give explicit ratings to the resources.
- Efficient privacy-preserving similar document detection92
This work proposes novel protocols to detect similar documents between two entities where documents cannot be openly shared with each other, and investigates a clustering based approach that significantly reduces the running time and achieves over 90% of accuracy in experiments.
- 92
- 92
- A Unified Syntax-aware Framework for Semantic Role Labeling91
This work extends existing models and proposes a unified framework to comprehensively explore the role of syntax for SRL task, and confirms that high-quality syntactic parse could still effectively enhance syntactically-driven SRL.
- 89
- Discriminative models of integrating document evidence and document-candidate associations for expert search84
Compared with the state-of-the-art language models for expert search, the proposed research can naturally integrate various document evidence and document-candidate associations into a single model without extra modeling assumptions or effort.
- Using sampled data and regression to merge search engine results84
This paper addresses the problem of merging results obtained from different databases and search engines in a distributed information retrieval environment and shows that the problem in uncooperative environments is simpler when viewed as a component of a distributed IR system that uses query-based sampling to create resource descriptions.
- PALM: Pre-training an Autoencoding&Autoregressive Language Model for Context-conditioned Generation79
PALM is presented with a novel scheme that jointly pre-trains an autoencoding and autoregressive language model on a large unlabeled corpus, specifically designed for generating new text conditioned on context.
- Multi-view transfer learning with a large margin approach78
This paper proposes a general framework: Multi-View Transfer Learning with a Large Margin Approach (MVTL-LM), and proposes an efficient optimization method, which is guaranteed to converge to ε precision in O(1/ε) steps.
- Unified utility maximization framework for resource selection77
This new framework shows an efficient and effective way to infer the probabilities of relevance of all the documents across the text databases and provides a more solid framework for distributed information retrieval.
- Preference preserving hashing for efficient recommendation74
Experiments show that the recommendation speed of the proposed PPH algorithm can be hundreds of times faster than original MF with real-valued features, and the recommendation accuracy is significantly better than previous work of hashing for recommendation.
- Non-Negative Matrix Factorization Clustering on Multiple Manifolds74
This paper proposes a novel clustering algorithm that explicitly models the intrinsic geometrical structure of the data on multiple manifolds with NMF and shows the advantage of the proposed algorithm on two real world datasets.
- A probabilistic graphical model for joint answer ranking in question answering74
A probabilistic graphical model is applied for answer ranking in question answering which estimates the joint probability of correctness of all answer candidates, from which the probability of Correctness of an individual candidate can be inferred.
- The FedLemur project: Federated search in the real world74
A prototype federated search system developed for the U.S. government's FedStats Web portal is discussed, and a series of experiments explore how well prior research results, parameter settings, and heuristics apply in the FedStats environment.
- Aspect Sentiment Classification with Document-level Sentiment Preference Modeling71
A Cooperative Graph Attention Networks (CoGAN) approach for cooperatively learning the aspect-related sentence representation and the importance of the document-level sentiment preference information to ASC is justified and the effectiveness of the approach capturing such information is demonstrated.
- Unified Dialog Model Pre-training for Task-Oriented Dialog Understanding and Generation69
SPACE is proposed, a novel unified pre-trained dialog model learning from large-scale dialog corpora with limited annotations, which can be effectively fine-tuned on a wide range of downstream dialog tasks.
- 69
- Decoupling Knowledge from Memorization: Retrieval-augmented Prompt Learning68
RetroPrompt is developed with the motivation of decoupling knowledge from memorization to help the model strike a balance between generalization and memorization and can reduce the reliance of language models on memorization; thus, improving generalization for downstream tasks.
- 68
- Rumor Detection on Social Media: Datasets, Methods and Opportunities67
An overview of the recent studies in the rumor detection field is given, a comprehensive list of datasets used for rumor detection is provided, and the important studies based on what types of information they exploit and the approaches they take are reviewed.
- t-Plausibility: Generalizing Words to Desensitize Text67
A novel information theoretic approach to text sanitization is provided and efficient heuristics to sanitize text documents are developed to protect sensitive and identifying information.
- 65
- Knowledge Transfer and Opinion Detection in the TREC2006 Blog Track65
The paper describes the opinion detection system developed in Carnegie Mellon University for TREC 2006 Blog track, which performed a two-stage process: passage retrieval and opinion detection.
- Latent Discriminative Models for Social Emotion Detection with Emotional Dependency62
The objective of which is to identify the evoked emotions of readers by online documents such as news articles, and a novel Latent Discriminative Model (LDM) is proposed for this task, which gives rise to a new Emotional Dependency-based LDM (eLDM).
- 61
- An Intelligent Tutor-Assisted Mathematics Intervention Program for Students With Learning Difficulties60
The purpose of this study was to explore the potential effects of the Please Go Bring Me-Conceptual Model-Based Problem Solving (PGBM-COMPS) intelligent tutor program on enhancing the multiplicative problem-solving skills of students with learning disabilities or difficulties in mathematics.
- Collaborative image retrieval via regularized metric learning60
This paper presents a novel metric learning approach, named “regularized metric learning,” for collaborative image retrieval, which learns a distance metric by exploring the correlation between low-level image features and the log data of users' relevance judgments.
- APE: Argument Pair Extraction from Peer Review and Rebuttal via Multi-task Learning58
This paper introduces a new argument pair extraction (APE) task on peer review and rebuttal in order to study the contents, the structure and the connections between them, and proposes a multitask learning framework based on hierarchical LSTM networks.
- Binary Codes Embedding for Fast Image Tagging with Incomplete Labels58
This paper proposes a novel Binary Codes Embedding approach for Fast Image Tagging (BCE-FIT) with incomplete labels, and constructs compact binary codes for both image examples and tags such that the observed tags are consistent with the constructed binary codes.
- 58
- 58
- “Bilingual Expert” Can Find Translation Errors57
This work proposes a general framework for automatic evaluation of translation output for most WMT quality evaluation tasks, and achieves the state-of-the-art performance in the quality estimation track of WMT 2017/2018.
- Semantic hashing using tags and topic modeling57
Experimental results indicate that the modeling of tag information and utilizing topic modeling are beneficial for improving the effectiveness of hashing separately, while the combination of these two techniques in the unified framework obtains even better results.
- A study of methods for normalizing user ratings in collaborative filtering57
The experiment results have shown that the decoupling method for rating normalization is more effective than the Gaussian normalization method in improving the performance of collaborative filtering algorithms.
- Proton: Probing Schema Linking Information from Pre-trained Language Models for Text-to-SQL Parsing56
A novel framework to elicit relational structures from large-scale pre-trained language models (PLMs) via a probing procedure based on Poincaré distance metric, and use the induced relations to augment current graph-based parsers for better schema linking is proposed.
- VECO: Variable and Flexible Cross-lingual Pre-training for Language Understanding and Generation56
A cross-attention module is plugged into the Transformer encoder to explicitly build the interdependence between languages and outperforms all existing cross-lingual models and state-of-the-art Transformer variants on WMT14 English-to-German and English- to-French translation datasets.
- Preview, Attend and Review: Schema-Aware Curriculum Learning for Multi-Domain Dialogue State Tracking56
This paper proposes a model-agnostic framework called Schema-aware Curriculum Learning for Dialog State Tracking (SaCLog), which consists of a preview module that pre-trains a DST model with schema information, a curriculum module that optimizes the model with CL, and a review module that augments mispredicted data to reinforce the CL training.
- MReD: A Meta-Review Dataset for Structure-Controllable Text Generation53
This paper introduces a new text generation dataset, named MReD, which consists of 7,089 meta-reviews and all its 45k meta- review sentences are manually annotated with one of the 9 carefully defined categories, including abstract, strength, decision, etc.
- Ranking preserving hashing for fast similarity search53
This paper proposes a novel Ranking Preserving Hashing (RPH) approach that directly optimizes a popular ranking measure, Normalized Discounted Cumulative Gain (NDCG), to obtain effective hashing codes with high ranking accuracy.
- Good Visual Guidance Makes A Better Extractor: Hierarchical Visual Prefix for Multimodal Entity and Relation Extraction52
A novel Hierarchical Visual Prefix fusion NeTwork (HVPNeT) for visual-enhanced entity and relation extraction, aiming to achieve more effective and robust performance.
- Dynamic Hybrid Relation Exploration Network for Cross-Domain Context-Dependent Semantic Parsing52
This paper presents a dynamic graph framework that is capable of effectively modelling contextual utterances, tokens, database schemas, and their complicated interaction as the conversation proceeds, and employs a dynamic memory decay mechanism that incorporates inductive bias to integrate enriched contextual relation representation.
- eventAI at SemEval-2019 Task 7: Rumor Detection on Social Media by Exploiting Content, User Credibility and Propagation Information52
This paper describes the system for SemEval 2019 RumorEval: Determining rumor veracity and support for rumors, which includes neural network models as well as the traditional classification algorithms.
- Relation Extraction as Open-book Examination50
This work regards RE as an open-book examination and proposes a new semiparametric paradigm of retrieval-enhanced prompt tuning for relation extraction, which can achieve state-of-the-art in both standard supervised and few-shot settings.
- GlobalWoZ: Globalizing MultiWoZ to Develop Multilingual Task-Oriented Dialogue Systems47
A novel data curation method is introduced that generates GlobalWoZ — a large-scale multilingual ToD datasets globalized from an English ToD dataset for three unexplored use cases of mult bilingual ToD systems.
- Syntax-Aware Neural Semantic Role Labeling47
Experiments on the benchmark CoNLL-2005 dataset show that syntax-aware SRL approaches can effectively improve performance over a strong baseline with external word representations from ELMo, and investigate several previous approaches for encoding syntactic trees to make a thorough study on whether extrantax-aware representations are beneficial for neural SRL models.
- 46
- Multiple Instance Learning on Structured Data46
This paper explores the research problem as multiple instance learning on structured data (MILSD) and formulates a novel framework that considers additional structure information and has the nice convergence property, with specified precision on each set of constraints.
- DialogueCSE: Dialogue-based Contrastive Learning of Sentence Embeddings45
The proposed DialogueCSE introduces a novel matching-guided embedding mechanism, which generates a context-aware embedding for each candidate response embedding according to the guidance of the multi-turn context-response matching matrices and minimizes the contrastive loss across all pairs.
- A Multi-Task Learning Approach for Improving Product Title Compression with User Search Log Data45
A novel multi-task learning approach for improving product title compression with user search log data by utilizing a pointer network-based sequence-to-sequence approach and an attentive encoder-decoder approach for generating user search queries.
- Author disambiguation by hierarchical agglomerative clustering with adaptive stopping criterion45
A novel Hierarchical Agglomerative Clustering approach with Adaptive Stopping Criterion (HACASC) is proposed to adaptively cluster a set of publications that share a same ANS to individual clusters of publications with different author identities.
- A Probabilistic Framework for Answer Selection in Question Answering45
This work focuses on developing a unified framework that not only uses multiple resources for validating answer candidates, but also considers evidence of similarity among answer candidates in order to boost the ranking of the correct answer.
- Duplex Conversation: Towards Human-like Interaction in Spoken Dialogue Systems44
The concept of full-duplex in telecommunication is used to demonstrate what a human-like interactive experience should be and how to achieve smooth turn-taking through three subtasks: user state detection, backchannel selection, and barge-in detection.
- Semi-supervised Domain Adaptation for Dependency Parsing43
A simple domain embedding approach is proposed to merge the source- and target-domain training data, which is shown to be more effective than both direct corpus concatenation and multi-task learning and shows that a simple fine-tuning procedure can further boost cross-domain parsing accuracy by large margin.
- Sentiment Classification in Customer Service Dialogue with Topic-Aware Multi-Task Learning42
This study proposes a topic-aware multi-task learning (TML) approach which learns topic-enriched utterance representations in customer service dialogue by capturing various kinds of topic information.
- Self-attentive Biaffine Dependency Parsing42
This work for the first time applies the self-attention mechanism to dependency parsing as the replacement of BiLSTM, leading to competitive performance on both English and Chinese benchmark data.
- Aspect Sentiment Classification Towards Question-Answering with Reinforced Bidirectional Attention Network41
A Reinforced Bidirectional Attention Network (RBAN) approach is proposed to address two inherent challenges in ASC-QA, i.e., semantic matching between question and answer, and data noise.
- IDST at TREC 2019 Deep Learning Track: Deep Cascade Ranking with Generation-based Document Expansion and Pre-trained Language Modeling.40
This paper proposes a two-stage cascade ranking pipeline by taking the advantages of sequence-to-sequence generation and pre-trained language modeling, and proposes a query generation method for document expansion based on the pointer-generator model.
- 40
- The impact of a conceptual model-based mathematics computer tutor on multiplicative reasoning and problem-solving of students with learning disabilities39
Qualitative and quantities data from this study indicate that, in addition to nurturing fundamental concept of composite units, it is necessary to help students to understand underlying problem structures and move toward mathematical model-based problem representation and solving for generalized problem solving skills.
- Learning to hash on partial multi-modal data39
A novel hashing approach to deal with Partial Multi-Modal data is presented, in which the hashing codes are learned by simultaneously ensuring the data consistency among different modalities via latent subspace learning, and preserving data similarity within the same modality through graph Laplacian.
- Learning to Hash with Partial Tags: Exploring Correlation between Tags and Hashing Bits for Large Scale Image Retrieval39
A novel semi-supervised tag hashing (SSTH) approach that fully incorporates tag information into learning effective hashing function by exploring the correlation between tags and hashing bits and improves the effectiveness of hashing function through orthogonal transformation by minimizing the quantization error.
- M 3 IC: maximum margin multiple instance clustering39
This paper formulates a novel Maximum Margin Multiple Instance Clustering (M3IC) problem, which enables an efficient optimization solution with a combination of Constrained Concave-Convex Procedure (CCCP) and the Cutting Plane method.
- 38
- Alibaba Submission for WMT18 Quality Estimation Task38
The QE Brain system is presented, which proposes the neural Bilingual Expert model as a feature extractor based on conditional target language model with a bidirectional transformer and then processes the semantic representations of source and the translation output with a Bi-LSTM predictive model for automatic quality estimation.
- 37
- Document clustering with universum37
A novel document clustering technique -- Document Clustering with Universum, which utilizes the Universum examples to improve the clustering performance and an extensive set of experiments are conducted to demonstrate the effectiveness and efficiency of the proposed algorithm.
- Discriminative probabilistic models for passage based retrieval37
This paper proposes two probabilistic models to formally model the evidence of a set of top ranked passages in a document and presents a combination approach to combine the ranked lists of document retrieval and passage-based retrieval.
- Alibaba at IJCNLP-2017 Task 1: Embedding Grammatical Features into LSTMs for Chinese Grammatical Error Diagnosis Task.36
Alibaba NLP team system on IJCNLP 2017 shared task No. 1 Chinese Grammatical Error Diagnosis is introduced, mainly based on the LSTM-CRF model and 3 ensemble strategies are applied to improve the performance.
- 35
- Boosting performance of bio-entity recognition by combining results from multiple systems35
Several Meta biomedical named-entity recognition algorithms that combine recognition results of various recognition systems are proposed that can be improved from 0.72, which is attained by the best individual system, to 0.96 by the Meta entity recognition approach.
- 33
- A joint probabilistic classification model for resource selection32
A joint probabilistic classification model for resource selection that estimates the probability of relevance of information sources in a joint manner by considering both the evidence of individual sources and their relationship is proposed.
- Initial results of using an intelligent tutoring system with Alice31
This paper focuses on the development of several tutorials for teaching specific introductory programming concepts that have been created using stencils and initial results concerning usability and effectiveness of these stencil-based tutorials are provided.
- VECO: Variable Encoder-decoder Pre-training for Cross-lingual Understanding and Generation30
A variable encoder-decoder (VECO) pre-training approach to unify the two mainstreams in both model architectures and pre- training tasks, which delivers new state-of-the-art results on various cross-lingual understanding tasks of the XTREME benchmark.
- Privacy-preserving IR: when information retrieval meets privacy and security30
This privacy-preserving IR workshop aims to spur research that brings together the research fields of IR and privacy/security, and research that mitigates privacy threats in information retrieval by constructing novel algorithms and tools that enable web users to better understand associated privacy risks.
- Efficient Pig Counting in Crowds with Keypoints Tracking and Spatial-aware Temporal Response Filtering29
This work presented a real-time automated pig counting system in crowds using only one monocular fisheye camera with an inspection robot, and showed that it achieved performance superior to human.
- DM_NLP at SemEval-2018 Task 12: A Pipeline System for Toponym Resolution29
DM-NLP’s system was developed for toponym detection, disambiguation and end-to-end resolution which is a pipeline of the former two and achieved the best performance among all the submitted results in each sub task.
- 29
- Unified filtering by combining collaborative filtering and content-based filtering via mixture model and exponential model29
A unified probabilistic framework that allows the mutual interaction between content information and rating information in an integrated way and outperforms a pure collaborative filtering approach, a pure content-based filtering approach and another unified filtering algorithm is presented.
- 28
- Sentiment Classification towards Question-Answering with Hierarchical Matching Network28
A high-quality annotated corpus with specially-designed annotation guidelines for QA-style sentiment classification is created and a three-stage hierarchical matching network is proposed to explore deep sentiment information in a QA text pair.
- Bi-VLDoc: bidirectional vision-language modeling for visually-rich document understanding27
This work investigates the problem of vision-language joint representation learning for VrDU mainly from the perspective of supervisory signals and proposes a pre-training paradigm called Bi-VLDoc, in which a bidirectional vision-language supervision strategy and a vision-language hybrid-attention mechanism are devised to learn stronger cross-modal document representations with richer semantics.
- A Dataset for Hyper-Relational Extraction and a Cube-Filling Approach27
This work proposes CubeRE, a cube-filling model inspired by table- filling approaches and explicitly considers the interaction between relation triplets and qualifiers, and further proposes acube-pruning method to improve model scalability and reduce negative class imbalance.
- Achieving Human Parity on Visual Question Answering27
A novel hierarchical integration of vision and language AliceMind-MMU (ALIbaba’s Collection of Encoder-decoders from Machine IntelligeNce lab of Damo academy - MultiMedia Understanding) is introduced, which leads to similar or even slightly better results than a human being does on VQA.
- 27
- A Unified Span-Based Approach for Opinion Mining with Syntactic Constituents27
The experimental results show that the proposed unified span-based approach achieves significant improvements over previous works in the exact F1 score and reduces the number of wrongly-predicted opinion expressions and roles, showing the effectiveness of the method.
- Argument Pair Extraction via Attention-guided Multi-Layer Multi-Cross Encoding27
A novel attention-guided multi-layer multi-cross encoding scheme to address the challenges of argument pair extraction and an auxiliary attention loss is introduced to guide each argument to align to its paired argument.
- Entity Retrieval with Hierarchical Relevance Model, Exploiting the Structure of Tables and Learning Homepage Classifiers.27
An overview of the work done for the TREC 2009 Entity track is given, which proposes a hierarchical relevance retrieval model for entity ranking and exploits the structure of tables and lists to identify the target entities from them.
- Learning from past queries for resource selection27
This paper proposes a new resource selection technique (which is called qSim) that utilizes the search results of past queries for estimating the utilities of available information sources for a specific user query.
- 26
- Cross-domain and cross-category emotion tagging for comments of online news26
An extensive set of experimental results on four datasets from popular online news services demonstrates the effectiveness of the proposed models in cross-domain emotion tagging for comments of online news in both the scenarios of sharing the same emotion categories or having different categories in the source and target domains.
- Identifying similar people in professional social networks with discriminative probabilistic models26
This paper proposes a discriminative probabilistic model that identifies latent content and graph classes for people with similar profile content and social graph similarity patterns, and learns a specialized similarity model for each latent class.
- Effective query generation and postprocessing strategies for prior art patent search26
An automated query generation and postprocessing method for prior art patent search is proposed that first constructs structured queries by combining terms extracted from different fields of a query patent and then reranks the retrieved patents by utilizing the International Patent Classification (IPC) code similarities between the query Patent and the retrieving patents along with the retrieval score.
- Discriminative probabilistic models for expert search in heterogeneous information sources26
This work investigates how to merge and weight heterogeneous knowledge sources in the context of expert finding and proposes a series of discriminative probabilistic models which have increasing capability to associate the combination weights with specific experts and queries.
- Combining evidence with a probabilistic framework for answer ranking and answer merging in question answering26
A unified probabilistic framework which combines multiple evidence to address challenges in answer ranking and answer merging and significantly improves performance of a QA system is proposed.
- CLEF 2005: Multilingual Retrieval by Combining Multiple Multilingual Ranked Lists26
Experimental results demonstrate the advantage of the query-specific and language-specific models against several other alternatives in multi-8 two-years-on retrieval and results merging.
- 26
- The Effect of Database Size Distribution on Resource Selection Algorithms25
Extended versions of two well-known resource selection algorithms are proposed: CORI and KL divergence in order to consider the factors of database size distributions, and compare them with the lately proposed Relevant Document Distribution Estimation (ReDDE) resource selection algorithm.
- Human-Like Decision Making: Document-level Aspect Sentiment Classification via Hierarchical Reinforcement Learning24
This paper proposes a new Hierarchical Reinforcement Learning (HRL) approach to DASC that incorporates clause selection and word selection strategies to tackle the data noise problem in the task of DASC.
- Perceive Your Users in Depth: Learning Universal User Representations from Multiple E-commerce Tasks24
- Learning for Efficient Supervised Query Expansion via Two-stage Feature Selection24
It is pointed out that the cost of SQE mainly comes from term feature extraction, and a Two-stage Feature Selection framework (TFS) is proposed to address this problem, which can substantially reduce the time cost for SQE, while maintaining its effectiveness.
- User Comment Analysis for Android apps and CSPI Detection with Comment Expansion.24
A supervised multi-label learning method utilizing comment expansion is adopted to detect different types of CSPI described by this label system, and experiments shows that the proposed method outperforms the method without the comment expansion.
- Probabilistic Answer Ranking Models for Multilingual QA: Independent Prediction Model and Joint Prediction Model24
Two probabilistic models for answering ranking in the multilingual question-answering (QA) task, which finds exact answers to a natural language question written in different languages, are presented.
- Probabilistic models for answer-ranking in multilingual question-answering24
Two probabilistic models for answering ranking in the multilingual question-answering (QA) task, which finds exact answers to a natural language question written in different languages, are presented.
- ENT-DESC: Entity Description Generation by Exploring Knowledge Graph23
A large-scale and challenging dataset that involves retrieving abundant knowledge of various types of main entities from a large knowledge graph (KG), which makes the current graph-to-sequence models severely suffer from the problems of information loss and parameter explosion while generating the descriptions.
- Detect Camouflaged Spam Content via StoneSkipping: Graph and Text Joint Embedding for Chinese Character Variation Representation23
A novel framework to jointly model Chinese variational, semantic, and contextualized representations for Chinese text spam detection task with a Variation Family-enhanced Graph Embedding algorithm designed based on a Chinese character variation graph is proposed.
- Supervised Treebank Conversion: Data and Approaches23
This work proposes two simple yet effective conversion approaches (pattern embedding and treeLSTM) based on the state-of-the-art deep biaffine parser that are superior to the widely used multi-task learning framework in multi-treebank exploitation and leads to significantly higher parsing accuracy.
- Weighted hashing for fast large scale similarity search23
A novel method, named Weighted Hashing (WeiHash), to assign different weights to different hashing bits, by simultaneously preserving the similarity between data examples and balancing the variance of each hashing bit is proposed.
- Mixture model with multiple centralized retrieval algorithms for result merging in federated search23
A mixture probabilistic model is proposed to learn more appropriate combination weights with respect to different types of information sources with some training data to deal with heterogeneous information sources.
- 23
- Learning to Identify Students' Off-Task Behavior in Intelligent Tutoring Systems23
A machine learning model that can automatically identify off-task behaviors of students while using an intelligent tutoring system is proposed and a robust Ridge Regression algorithm is designed to estimate model parameters.
- An effective and efficient results merging strategy for multilingual information retrieval in federated search environments23
This paper proposes an effective and efficient approach for the results merging task of multilingual ranked lists that downloads only a small number of documents from the individual ranked lists of each user query to calculate comparable document scores by utilizing both the query-based translation method and the document-basedtranslation method.
- Adjusting Mixture Weights of Gaussian Mixture Model via Regularized Probabilistic Latent Semantic Analysis23
A regularized probabilistic latent semantic analysis model (RPLSA), which can properly adjust the amount of model flexibility so that not only the training data can be fit well but also the model is robust to avoid the overfitting problem.
- Overview of the NLPCC 2019 Shared Task: Cross-Domain Dependency Parsing22
An overview of the NLPCC 2019 shared task on cross-domain dependency parsing, including (1) the data annotation process, (2) task settings, methods, results, and analysis of submitted systems and the authors' recent work (Li+19), and discussions on related works and future directions.
- Legal Intelligence for E-commerce21
An extensive set of experiments with a large dispute case dataset collected from a world leading e-commerce platform shows that the proposed model can more accurately characterize a dispute case through buyer, seller, and transaction viewpoints for legal judgment prediction against several alternatives.
- Syntax-Enhanced Self-Attention-Based Semantic Role Labeling21
The experiment results demonstrate that with proper incorporation of the high quality syntactic information, the model achieves a new state-of-the-art performance for the Chinese SRL task on the CoNLL-2009 dataset.
- 21
- 21
- 21
- A Discriminative Data-Dependent Mixture-Model Approach for Multiple Instance Learning in Image Classification21
The new MM-MIL approach explicitly handles the multi-target problem by introducing a data-dependent mixture model, which allows positive instances to come from different clusters in a flexible manner and allows effective and efficient learning in high dimensional feature space.
- Review-based Question Generation with Adaptive Instance Transfer and Augmentation20
This work proposes to explore question generation as a new way of review information exploitation, namely generating questions that can be answered by the corresponding review sentences, with an iterative learning framework with adaptive instance transfer and augmentation.
- MILEAGE: Multiple Instance LEArning with Global Embedding20
A novel framework is proposed, in which the global feature vectors for traditional learning methods are integrated into the MIL setting, and a large margin method is formulated to adaptively tune the weights on the two different kinds of feature representations for each example and trains the classifier simultaneously.
- 19
- 18
- Enhancing Multilingual Language Model with Massive Multilingual Knowledge Triples18
This work explores methods to make better use of the multilingual annotation and language agnostic property of KG triples, and presents novel knowledge based multilingual language models (KMLMs) trained directly on the knowledge triples.
- Mobile App Security Risk Assessment: A Crowdsourcing Ranking Approach from User Comments18
A novel perspective to view the risk assessment of an app from its user comments as a crowdsourcing problem and adopt ranking model as the evaluation method is provided.
- Search result diversification in resource selection for federated search18
A learning based approach to combine multiple resource selection algorithms for result diversification, which can further improve the performance and a set of new metrics for resource selection in federated search to evaluate the diversification performance of different approaches.
- 18
- Unsupervised Learning of Deterministic Dialogue Structure with Edge-Enhanced Graph Auto-Encoder17
A conversational graph (CG) is proposed to represent deterministic dialogue structure where nodes and edges represent the utterance and context information respectively and a self-supervised objective is introduced with the response selection task to guide the unsupervised learning of the dialogue structure.
- Sexual Harassment Story Classification and Key Information Identification17
This study proposed neural network models to extract key elements including harasser, time, location and trigger words from Safecity stories, and demonstrated that encoding the key element information in the story categorization model can improve its performance.
- Unsupervised Learning Helps Supervised Neural Word Segmentation17
This work makes the first attempt at exploring adding unsupervised segmentation information into neural supervised segmenter, and verifies the proposed approach outperforms the widelyused pre-training approach in terms of effectively making use of freely abundant unlabeled data.
- Adaptive Knowledge Transfer for Multiple Instance Learning in Image Classification17
An extensive set of experimental results demonstrate that the proposed AKT-MIL approach substantially outperforms several state-of-the-art algorithms on two benchmark datasets, especially in the scenario when very few training examples are available in the target domain.
- Learning compact hashing codes for efficient tag completion and prediction17
A novel efficient Hashing approach for Tag Completion and Prediction (HashTCP) is proposed that can achieve similar or even better accuracy with state-of-the-art methods and can be much more efficient, which is important for large scale applications.
- 16
- An Entity Class-Dependent Discriminative Mixture Model for Cumulative Citation Recommendation16
A novel entity class-dependent discriminative mixture model is proposed by introducing a latent entity class layer to model the correlations between entities and latent entity classes to achieve better performance when dealing with a broad range of entities.
- 16
- Analysis of an expert search query log16
It is found that expert search users generally issue shorter queries, more common queries, and use more advanced search features, with fewer queries in a session, than general Web search users do.
- Predicting Correctness of Problem Solving from Low-Level Log Data in Intelligent Tutoring Systems16
A learning based metho d that can automatically determine how likely a student is to give a correct answer to a problem in an intelligent tutoring system is proposed.
- 15
- Forecasting user visits for online display advertising15
An extensive set of experiments along with detailed analysis carried out on real-world data from Yahoo! demonstrates the advantage of the proposed latent class models in forecasting online user visits in online display advertising.
- Purdue at TREC 2010 Entity Track: a Probabilistic Framework for Matching Types between Candidate and Target Entities15
This paper proposes a unified probabilistic framework by incorporating the matching between target entity types and candidate entity types for both the Related Entity Finding (REF) task and the Entity List Completion (ELC) task in this track.
- Federated search of text search engines in uncooperative environments15
This paper focuses on hidden information, which is very valuable, and can only be accessed via an alternative search model other than the centralized retrieval model used by conventional search engines.
- Combining Multiple Resources, Evidences and Criteria for Genomic Information Retrieval15
The query expansion method utilizes multiple external biomedical resources to extract acronyms, aliases, and synonyms, and a post-processing step which combines the evidence from multiple scoring methods to improve relevance-based passage rankings is proposed.
- Recommending Complementary Products in E-Commerce Push Notifications with a Mixture Model Approach14
A mixture model approach for predicting push message open rate for a post-purchase complementary product recommendation task and the results show that the proposed method is superior than several existing solutions by a significant margin.
- Probabilistic latent class models for predicting student performance14
A set of novel probabilistic latent class models for predicting student performance and using them to identify hidden classes of students and educational content that share similar characteristics are proposed.
- Sentiment detection with auxiliary data14
A novel method to model the distribution difference between different domains in sentiment detection by directly modeling the underlying joint distributions for different domains is proposed.
- Machine learning for information retrieval14
This tutorial will benefit a large body of audience in the information retrieval community, ranging from students who are new to machine learning to the seasoned researchers who would like to understand the recent advance in machine learning for information retrieval research.
- Predicting Correctness of Problem Solving in ITS with a Temporal Collaborative Filtering Approach14
This paper proposes a temporal CF approach to automatically predict the correctness of students' problem solving in an intelligent math tutoring system and shows that using multiple-interactions significantly improves the prediction accuracy while using sliding-time-windows doesn't make a significant difference.
- Automatic Text Categorization of Mathematical Word Problems14
A probabilistic meta classifier is proposed to combine the weighted results of two SVM classifiers with different word problem representations generated by different text preprocessing techniques to improve the categorization accuracy of text categorization for mathematical word problems.
- 13
- 13
- Knowledge-aware Named Entity Recognition with Alleviating Heterogeneity13
A new KB-aware NER framework (KaNa), which utilizes type-heterogeneous knowledge to improve NER, and demonstrates KaNa's state-of-the-art performance on five public benchmark datasets from different domains.
- A Research Agenda for the Study of Entropic Social Structural Evolution, Functional Roles, Adhocratic Leadership Styles, and Credibility in Online Organizations and Knowledge Markets13
The new social media enabled by the Internet and the Web have deeply changed the ways in which individuals interact and how knowledge is created and exchanged, which is opening up interesting new research questions for social science.
- Multi-Turn Dialogue Generation in E-Commerce Platform with the Context of Historical Dialogue12
A novel and extensible dialogue generation method by leveraging sellers’ historical dialogue information, which can be both accessible and informative, is proposed by utilizing innovative historical dialogue representation learning and historical dialogue selection mechanism.
- Improve Neural Entity Recognition via Multi-Task Data Selection and Constrained Decoding12
An entity recognition system that improves this neural architecture with two novel techniques, including Multi-Task Data Selection, which ensures the consistency of data distribution and labeling guidelines between source and target datasets and constrained decoding using knowledge base.
- A latent pairwise preference learning approach for recommendation from implicit feedback12
This paper proposes a latent pairwise preference learning (LPPL) approach for recommendation with implicit feedback and conducts experiments on the testbed from a real-world recommender system to demonstrate that the proposed approach can effectively improve the recommendation performance against several baseline algorithms.
- 12
- FacFinder: Search for Expertise in Academic Institutions*12
Finder is presented as a publicly accessible faculty expertise search and ranking system across multiple universities and the results of experiments aimed at evaluating the effectiveness and efficiency of FacFinder in its operational environment are presented.
- 12
- Protecting source privacy in federated search12
This paper proposes a two-step federated search protocol that protects the privacy of information sources and is the first attempt to address the research problem of protecting source privacy in federated text search.
- SUN: Exploring Intrinsic Uncertainties in Text-to-SQL Parsers11
A model uncertainty constraint is proposed to refine the query representations by enforcing the output representations of different perturbed encoding networks to be consistent with each other and improve the generalizability and stability of neural text-to-SQL parsers.
- 11
- Learning to Hash on Structured Data11
This paper improves the effectiveness of hashing function through orthogonal transformation by minimizing the quantization error and formulates anovel framework that considers additional structure information in hashing techniques.
- A weighted curve fitting method for result merging in federated search11
A weighted curve fitting method for result merging in federated search that explicitly models the importance of information from overlapping documents over non-overlapping ones and weights documents at different positions differently.
- Discriminative graphical models for faculty homepage discovery11
This paper views faculty homepage discovery as text categorization problems by utilizing Yahoo BOSS API to generate a small list of high-quality candidate homepages and proposes a conditional undirected graphical model to make joint predictions by capturing the dependence of the decisions on all the candidate pages.
- Automatic target recognition of radar HRRP based on high order central moments features11
A novel approach of feature extraction and dimension reduction based on extended high order central moments is proposed in order to reduce the dimension of range profiles and can achieve good performance of stability, shift independence and higher recognition rate.
- Exploration of the tradeoff between effectiveness and efficiency for results merging in federated search11
This paper proposes a utility based optimization method that enables the system to automatically decide on the desired number of training documents to download according to the user's need for effectiveness and efficiency.
- Finding Camouflaged Needle in a Haystack?10
A novel task to dynamically locate the pornographic products from very large product collections is proposed and the proposed model, BerryPIcking TRee MoDel (BIRD), utilizes both product textual content and buyers' seeking behavior information as berrypicking trees.
- NLP_HZ at SemEval-2018 Task 9: a Nearest Neighbor Approach10
This paper proposes a simple but effective method for the discovery of hypernym sets based on word embedding, which can be used to measure the contextual similarities between words.
- 10
- Conceptual Model-Based Problem Solving That Facilitates Algebra Readiness: An Exploratory Study with Computer-Assisted Instruction9
The results indicate that the computer-assisted COMPS program seemed to benefit students with LP; it enhanced their problem-solving skills and promoted algebrareadiness.
- Emotion tagging for comments of online news by meta classification with heterogeneous information sources9
A novel Meta classification approach is proposed that exploits heterogeneous information sources such as the content of the comments and the emotion tags of news articles generated by users to predict emotions for the comments of online news.
- Serendipitous learning9
A novel notion of Serendipitous Learning (SL), which is defined to address the learning scenarios in which the label space can be enlarged during the testing phase, and a large margin approach is proposed to solve SL.
- Learning to Identify Students’ Relevant and Irrelevant Questions in a Micro-blogging Supported Classroom9
Results and analysis show that utilizing the correlation between questions and available lecture materials in a lecture along with personalization and question text leads to significantly higher categorization accuracy than i) using personalization along with question text and ii) using question text alone.
- Learning to Rank Documents for Ad-Hoc Retrieval with Regularized Models9
Two regularized models to estimate the query-dependent weights are proposed, one is a variant of EM algorithm - Deterministic Annealing EM (DAEM), and another is a L2- regularized log-linear model (RLM), both of them rely on some regularization methods to avoid unbalanced mixture weight and to make them better fit the test data.
- 9
- A Simple Concatenation can Effectively Improve Speech Translation8
Inspired by the works of video Transformer, a simple unified cross-modal ST method is proposed, which concatenates speech and text as the input, and builds a teacher that can utilize both cross- modal information simultaneously.
- From Cloze to Comprehension: Retrofitting Pre-trained Masked Language Model to Pre-trained Machine Reader8
Although conceptually simple, PMR is particularly effective in solving extraction tasks including Extractive Question Answering and Named Entity Recognition, where it shows tremendous improvements over previous approaches especially under low-resource settings.
- 8
- 8
- TweetSenti: Target-dependent Tweet Sentiment Analysis8
TweetSenti is a system for analyzing the sentiment of an entity in tweets and built a web-based application that users can interact with and analyze an entity's sentiment in Twitter at real-time.
- DM_NLP at SemEval-2018 Task 8: neural sequence labeling with linguistic features8
This paper describes the submissions for SemEval-2018 Task 8: Semantic Extraction from CybersecUrity REports using NLP, and describes the system solving Subtask 2, which relies on a recurrent neural network named BiLSTM-CNN-CRF with rich linguistic features.
- Learning Beyond the Predefined Label Space8
A novel notion of Serendipitous Learning (SL), which is defined to address the learning scenarios in which the label space can be enlarged during the testing phase, and a large margin approach is proposed to solve SL.
- 8
- Similar Trademark Detection via Semantic, Phonetic and Visual Similarity Information7
This paper proposes a neural network model that exploits the semantic, phonetic and visual similarities between two textual trademarks in Chinese, and shows that the proposed model outperforms other approaches.
- One vs. Many QA Matching with both Word-level and Sentence-level Attention Network7
A novel QA matching approach, namely One vs. Many Matching, which aims to address the novel scenario where one question sentence often has an answer with multiple sentences is proposed, and is improved by employing both word-level and sentence-level attentions.
- Detecting temporal patterns of user queries7
This article presents a novel classification scheme from the perspective of queries' temporal patterns and extracts three groups of features based on the queries' search volume time series and uses a support vector machine to automatically detect the temporal patterns of user queries.
- Related entity finding by unified probabilistic models7
The proposed unified probabilistic approach is applied to the TREC Entity List Completion task and also demonstrates good performance, indicating the benefits of explicitly modeling source entity in related entity finding.
- Forecasting counts of user visits for online display advertising with probabilistic latent class models7
This paper proposes a probabilistic latent class model to automatically learn the underlying user visit patterns among multiple Web pages and demonstrates the advantage of using latent classes in forecasting online user visits.
- Query Expansion and Message-passing Algorithms for TREC Microblog Track7
It is found that query expansion is a very useful technique for microblog retrieval, while anity propagation could achieve a comparable performance when combining with other techniques.
- A Joint Probabilistic Classification Model of Relevant and Irrelevant Sentences in Mathematical Word Problems7
Empirical results and analysis show that i) it is highly beneficial not to remove stopwords and ii) utilizing part of speech tagging does not make a significant improvement although it has been shown to be effective for the related task of math word problem type classification.
- Strategies for Effective Chemical Information Retrieval7
The methods developed for these two tasks, an automated query generation method that uses all title words, and selects sets of terms from the claims, abstract and description fields of query patents to transform a query patent into a search query, are described.
- Thresholding Strategies for Text Classifiers: TREC-2005 Biomedical Triage Task Experiments7
A thresholding method is described that combines the outputs from logistic regression and support vector machine by applying a joint thresholding strategy.
- Language model for IR using collection information7
A new language model is presented that is able to take advantage of the category information for documents to improve the retrieval accuracy and outperforms the traditional language model over the TREC4 dataset.
- 6
- Disentangled representation for sequential treatment effect estimation6
The experimental results indicate the powerful capacity of the model in learning disentangled representations from longitudinal observational data and dealing with the time-varying confounders, and demonstrate the surpassing performance achieved by the proposed model on dynamic treatment effect estimation.
- 6
- 6
- 6
- SIGIR 2017 Workshop on eCommerce (ECOM17)6
This workshop brings together researchers and practitioners of eCommerce IR to discuss topics unique to it, to set a research agenda, and to examine how to build a dataset for research into this fascinating topic.
- 6
- PIR 2014 The First International Workshop on Privacy-Preserving IR6
The First International Workshop on Privacy-Preserving IR: When Information Retrieval Meets Privacy and Security (PIR 2013) was held as part of the SIGIR 2014 conference in Gold Coast, Australia.
- Combining gene sequence similarity and textual information for gene function annotation in the literature6
Empirical studies on two testbeds demonstrate that the combination of sequence similarity scores, function prior probabilities and textual information improves the accuracy of gene function annotation in the literature.
- 5
- 5
- Chinese E-Romance: Analyzing and Visualizing 7.92 Million Alibaba Valentine's Day Purchases5
This exploratory study investigates the extent to which each sales category and each consumer group can exhibit romantic love within China's e-market throughout the two weeks leading up to 2019 Valentine's Day.
- Uncover Sexual Harassment Patterns from Personal Stories by Joint Key Element Extraction and Categorization5
This study manually annotated those stories with labels in the dimensions of location, time, and harassers’ characteristics, and marked the key elements related to these dimensions, and applied natural language processing technologies with joint learning schemes to automatically categorize these stories in those dimensions and extract key elements at the same time.
- Alibaba at IJCNLP-2017 Task 2: A Boosted Deep System for Dimensional Sentiment Analysis of Chinese Phrases.5
Team Alibaba’s systems participating IJCNLP 2017 shared task No. 2 Dimensional Sentiment Analysis for Chinese Phrases (DSAP) achieved 2nd in mean rank, and mainly utilize a multi-layer neural networks for model training.
- Sparse Semantic Hashing for Efficient Large Scale Similarity Search5
A unified framework is designed for ensuring the hidden semantic structure among the documents by a sparse coding model, while at the same time preserving the document similarity via graph Laplacian and an iterative coordinate descent procedure is proposed for solving the optimization problem.
- 5
- Using Past Queries for Resource Selection in Distributed Information Retrieval5
A new resource selection technique is proposed that utilizes the search results of past queries for estimating the utilities of available information sources for a specific user query and a combined resource selection approach is proposed to integrate the two approaches of learning from past queries and using static sampled information.
- 5
- A Bayesian framework for knowledge driven regression model in micro-array data analysis5
This paper presented a full Bayesian framework to effectively exploit the similarity information of the input variables for linear regression to address the sparse data problem in the linear regression model.
- 4
- 4
- 4
- 4
- LDTM: A Latent Document Type Model for Cumulative Citation Recommendation4
An extensive set of experiments has been conducted on TREC-KBA-2013 dataset, and the results demonstrate that this model can yield a significant performance gain in recommendation quality as compared to the state-of-the-art.
- 3
- Context-Adaptive Document-Level Neural Machine Translation3
A data-adaptive method is introduced that enables the model to adopt the necessary and helpful context and can significantly improve the performance over the previous methods with a gain up to 1.99 BLEU points.
- 3
- 3
- 3
- 3
- An Adversarial Joint Learning Model for Low-Resource Language Semantic Textual Similarity3
This work proposes to jointly learn the low-resource language STS task and that of a resource-rich one, which only relies on multilingual word embeddings, via a translation matrix and introduces a language discriminator introduced as an adversarial teacher.
- Improve Neural Mention Detection and Classification via Enforced Training and Inference Consistency.3
A multi-task version of BiLSTM-CRF model is applied to the NERC task, to better utilize additional data sources and to enforce data and label consistency at both training and prediction time.
- 3
- Supervised Local Contexts Aggregation for Effective Session Search3
This paper proposes Supervised Local Context Aggregation (SLCA) as a principled framework for complex session context modeling and shows that the proposed SLCA algorithm outperforms many other session search methods, and achieves the state-of-the-art results.
- 3
- 3
- 3
- 3
- Leveraging online shopping behaviors as a proxy for personal lifestyle choices: New insights into chronic disease prevention literacy2
Mining online shopping behaviors can point medical experts to a series of lifestyle issues associated with chronic diseases that are less explored to date, and hopefully, unobtrusive chronic disease surveillance via e-commerce sites can grant consenting individuals a privilege to be connected more readily with the medical profession and sophistication.
- Knowledge Graph Empowered Entity Description Generation.2
This paper introduces a large-scale and challenging dataset that involves exploring large knowledge graphs (KG) to retrieve abundant knowledge of various types of main entities, which makes the current graph-to-sequence models severely suffered from the problems of information loss and parameter explosion while generating the description text.
- Report on the SIGIR 2017 Workshop on eCommerce (ECOM17)2
The SIGIR 2017 Workshop on eCommerce (ECOM17), was a full day workshop that took place on Friday, August 11, 2017 in Tokyo, Japan to serve as a platform for publication and discussion of Information Retrieval and NLP research and their applications in the domain of eCommerce.
- Determining expert research areas with multi-instance learning of hierarchical multi-label classification model2
A novel approach, Multi-instance Learning of Hierarchical Multi-label Classification Model (MIHML) for the task, which effectively identifies multiple research areas in a hierarchy from individual documents within the profile of a researcher is proposed.
- Which Tweets Will Be Headlines? A Hierarchical Bayesian Model for Bridging Social Media and Traditional Media2
A Twitter Pick-Up Relational (TPUR) model is proposed to simultaneously integrate the dependence between the traditional media's interests towards a user and the content of each tweet, and the influence of each user are integrated in a hierarchical bayesian model.
- A robust one class Bayesian approach for masquerade detection2
A robust one-class Bayesian approach for masquerade detection that explicitly considers model uncertainty by integrating out the unknown model parameters for generating robust results, while previous one- class methods only use a single point estimate to find an optimal model.
- 1
- 1
- 1
- 1
- NAI-SEA at SemEval-2018 Task 5: An Event Search System1
Alibaba’s participating system in the semEval-2018 Task5: Counting Events and Participants in the Long Tail ranks first in all the three subtasks on document level retrieval, and it also ranks first on incident-level evaluation by RSME measure in subtask 3.
- Relevancy prediction of micro-blog questions in an educational setting.1
This work proposes a novel solution for predicting the relevancy of a question asked in a class by looking at the questions asked in previous semesters, the similarity of the question to the lecture material, as well as a set of question features such as the number of students’ votes, number of replies, and whether it was asked anonymously.
- 1
- Predicting Low-Quality Wikipedia Articles Using User’s Judgements1
This paper utilizes article ratings from Wikipedia users for the first time to assess article quality and defines “low-quality” based on those ratings and design automatic methods to identify potential low-quality articles.
- BIT and Purdue at TREC-KBA-CCR Track 20141
The results reveal that learning to rank approaches are more suitable for CCR with current evaluation methodology.
- Effective and efficient user and content modeling for intelligent tutoring systems1
Novel probabilistic approaches for effective and efficient user and content modeling for intelligent tutoring systems are proposed and tested on real-world datasets from the prototype system as well as on external large-scale datasets to demonstrate the effectiveness of the proposed models.
- 1
- Modeling search response time1
A novel query-specific and source-specific approach to model search response time is proposed by utilizing Ridge Regression and can be used to predict search response times for new queries.
- 1
- Detecting Salient Regions Based on "What" and "Where" Pathways of Visual Systems1
A novel model for salient region detection based on "what" and "where" pathways in an image is proposed and experimental results with natural images demonstrate its effectiveness and robustness.
- Conjugate Descent Algorithm of Independent Component Analysis1
Based on analysis of existing algorithms, conjugate gradient algorithm is given using minimal mutual information criteria and the simulation experiments prove its good convergence and robust.
- Discussion of the Data Compression Method Based on Singular Value Decomposition1
The paper goes deep to study three schemes of SVD_based image compression and proves the usage feasibility of Svd_basedimage compression.
- ACM WSDM 2023 Report–
This brief report provides an overview of WSDM 2023 with organization and program details and statistics from Conference Chairs and Program Committee Chairs, as well as a message from the W SDM Steering Committee Chair.
- Report on the 16th ACM International Conference on Web Search and Data Mining (WSDM 2023)–
This brief report provides an overview of WSDM 2023 with organization and program details and statistics from Conference Chairs and Program Committee Chairs, as well as a message from the W SDM Steering Committee Chair.
- –
- –
- –
- China's E-Commerce Romance: Analyzing and Visualizing 7.92 Million Alibaba Valentine's Day Purchases–
- Natural Language Technologies for Internet Applications–
This talk presents some recent research efforts and trends of four sets of NLP technologies for Internet applications and neural language model has been a very popular research direction in the last a few years and has significantly improved the performance of many applications.
- –
- –
- ProbKS: Keyword Search on Probabilistic Spatial Data–
This paper develops an IRTree-based Incremental Scoring Approach (ISA) that returns an iterator over tuples in decreasing order of text similarity and develops a parameterized probabilistic ranking algorithm, \(PRank^c\), that consumes the output of ISA interactively and calculates top-\(c\) confidence of these Tuples in linear time.
- Privacy-preserving and efficient friend recommendation in social networks–
This paper proposes two novel methods to recommend friends for a given user by using the common neighbors proximity measure in a privacy preserving manner, based on the properties of additive homomorphic encryption scheme and a universal hash function for efficiency purpose.
- Dealing with ambiguous and partial supervision in complex information retrieval applications–
This thesis investigates how to utilize different kinds of partial knowledge to alleviate the problem of ambiguity in different scenarios, which include the label ambiguity problem, the label incompleteness problem and some ambiguity problems in social media.
- Probabilistic approaches to entity retrieval–
This dissertation proposes formal probabilistic approaches to entity retrieval and present a discriminative learning framework for expert search, and presents a real-world expert search system for academic institutions: INdiana Database of University Research Expertise (INDURE).
- –
- –
- –
- Secure similar document detection–
This work proposes a novel scheme to detect similar documents between two entities where documents cannot be openly shared with each other, and presents experimental results to show the practical value of the proposed protocols.
- Research on an Instance-Based Workflow Engine for Dynamic Definition and Execution–
This paper puts forth an instance-based lightweight workflow engine for dynamic definition and execution that realizes the standard and individual modeling for non-structured business process, and enhances the flexible description and execution ability for non -structured processes.
- IBM Research Report On the Use of Hierachy Information in Mapping Patents to Biomedical Ontologies–
This paper argues that the ontology representing patents is very valuable since it represents the intellectual property owned by businesses and also is a means of associating chemical and biological entities that impact human health to most other manufactured goods and manufacturing processes.
- Study on the distribution of mast cell in the normal human uterus–
The relative reduction of MC number in the early secretory endometrium might benefit to the implantaton of blastocyst and the significant decrease in the menopausal endometricrium might be related with the high incidence of uterine malignomas in menopausal women.
- A Knowledge Increasable Method for Artificial Neural Network–
Experimental results show that such a system can effectively inherit the knowledge of its components and can gain novel knowledge gradually when new modules arrive, and the reliability, flexibility and scalability of the system are greatly improved.
- –
- Research for Simple Reliable Communication Protocol Communication on Computer Cluster–
This paper designed Linux SRCP, and applied this protocol to computer cluster and found that this protocol reduces the influence of system kernel.
- Design of CT Medical Image Display on Linux Platform–
The paper firstly introduces the structure and function of a concrete CT system, and then presents one of subsystem, designed and implemented by the authors, the image display subsystem_DISPLAY system, which concludes the defects and improvement scheme in the future.
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 2026-10-12. 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.