Large language models (LLMs) perform in-context learning (ICL) with minimal supervised examples, which benefits various natural language processing (NLP) tasks. One of the critical research focus is the selection of prompt demonstrations. Current approaches typically employ retrieval models to select the top-K most semantically similar examples as demonstrations. However, we argue that existing methods are limited since the label consistency is not guaranteed during demonstration selection. Our cognition derives from the Bayesian view of ICL and our rethinking of ICL from the transductive label propagation perspective. We treat ICL as a transductive learning method and incorporate latent concepts from Bayesian view and deduce that similar demonstrations guide the concepts of query, with consistent labels serving as estimates. Based on this understanding, we establish a label propagation framework to link label consistency with propagation error bounds. To model label consistency, we propose a data synthesis method, leveraging both semantic and label information, and use TopK sampling with Synthetic Data (TopK-SD) to acquire demonstrations with consistent labels. TopK-SD outperforms original TopK sampling on multiple benchmarks. Our work provides a new perspective for understanding the working mechanisms within ICL.

Fine-grained emotion classification (FEC) requires distinguishing subtly different emotions, where the dominant errors come from closely confusable categories.Recent progress relies on contrastive learning with hard-pair mining, implicitly assuming that a fixed similarity metric is sufficient to optimize informative pairs.We argue that this assumption is fragile because defining whether two utterances are similar becomes a problem when the label space is crowded, and hard-pair mining under a fixed metric can systematically miss the worst confusions.Thus, we treat the similarity function as a learnable component and design an adversarial metric learning (AML) framework.It follows theoretical interpretations of metric-robust representations that better separate confusable emotions.AML trains a pairwise discriminator to maximally confuse two targeted hard pair types, while training the encoder to remain discriminative under this worst-case learned metric.Our code and data are released on GitHub 1 .

Dense retrieval represents queries and documents as high-dimensional embeddings, but these representations can be redundant at the query level: for a given information need, only a subset of dimensions is consistently helpful for ranking.Prior work addresses this via pseudo-relevance feedback (PRF) based dimension importance estimation, which can produce query-aware masks without labeled data but often relies on noisy pseudo signals and heuristic test-time procedures.In contrast, supervised adapter methods leverage relevance labels to improve embedding quality, yet they learn global transformations shared across queries and do not explicitly model query-aware dimension importance.We propose a Query-Aware Adaptive Dimension Selection framework that learns to predict per-dimension importance directly from query embedding.We first construct oracle dimension importance distributions over embedding dimensions using supervised relevance labels, and then train a predictor to map a query embedding to these label-distilled importance scores.At inference, the predictor selects a query-aware subset of dimensions for similarity computation based solely on the query embedding, without pseudorelevance feedback.Experiments across multiple dense retrievers and benchmarks show that our learned dimension selector improves retrieval effectiveness over the full-dimensional baseline as well as PRF-based masking and supervised adapter baselines.

Generalized Category Discovery (GCD) is a critical task in open-world computing scenarios, aiming to automatically classify partially labeled data by recognizing both known and novel categories. However, existing GCD methods usually suffer from inherent bias toward known categories due to the exclusive pre-training on them and the absence of labeled data of novel categories. This bias can lead to significant misclassification and clustering errors for novel categories. Although recent approaches leverage pseudo-label training and contrastive learning to address this, they still lack explicit supervision to disentangle novel and known categories, resulting in performance bottlenecks. To address these limitations, we propose an Expectation-Maximization-driven Contrastive Disentanglement (EMCD) framework designed to explicitly disentangle novel and known categories. We particularly formulate the identification of novel categories as a latent variable estimation problem. Specifically, it incorporates an EM-disentangling regularization to softly identify novel category samples and a consistency regularization to enhance generalization. In addition, we leverage dual contrastive constraints, including a cluster-sample contrastive constraint and a sample-sample contrastive contrastive, to pull close samples of novel categories while pushing apart ambiguous samples near decision boundaries. Empirical results on 3 commonly used datasets demonstrate that our model is effective and outperforms previous state-of-the-art methods. Our code is available at https://github.com/YWY-only/EMCD.

Multimodal knowledge graph completion (MKGC) focuses on predicting missing head or tail entities in multimodal knowledge graphs. While previous studies have emphasized the integration of images and descriptive sentences to enhance entity representation learning, our research highlights that not all multimodal information is beneficial for MKGC. In fact, irrelevant multimodal data can lead to incorrect predictions and diminished performance. This finding underscores the need for a balance between leveraging relevant multimodal information and minimizing the impact of irrelevant data in knowledge graph representation learning. To address this challenge, we propose a self-adaptive fusion model for MKGC (SAFKGC). SAFKGC utilizes a cross-modal transformer with an innovative input sequence to select features and assess the importance of different modalities. It also identifies irrelevant multimodal information based on prediction confidence, weakening its influence while maintaining reasoning consistency between multimodal and unimodal models through soft-label learning. Experimental results across four widely used MKGC datasets demonstrate that our model achieves competitive or superior performance compared to entity-aware the state-of-the-art approaches. Additional experiments reveal that SAFKGC can dynamically adjust unimodal KGC results to enhance multimodal KGC outcomes. The code is available at https://anonymous.4open.science/r/SAFKGC-BBE4/ .

Supervised multimodal classification has been proven to outperform unimodal classification in the image-text domain. However, this task is highly dependent on abundant labeled data. To perform multimodal classification in data-insufficient scenarios, in this study, we explore semi-supervised multimodal classification (SSMC) that only requires a small amount of labeled data and plenty of unlabeled data. Specifically, we first design baseline SSMC models by combining known semi supervised pseudo-labeling methods with the two most commonly used modal fusion strategies, i.e. feature-level fusion and label-level aggregation. Based on our investigation and empirical study of the baselines, we discover two complementarities that may benefit SSMC if properly exploited: the predictions from different modalities (modal complementarity) and modal fusion strategies for pseudo-labeling (strategic complementarity). Therefore, we propose a Modal and Strategic Complementarity (MSC) framework for SSMC. Concretely, to exploit modal complementarity, we propose to learn reliability weights for the predictions from different modalities and refine the fusion scores. To learn from strategic complementarity, we introduce a dual KL divergence loss to guide the balance of quantity and quality of pseudo-labeled data selection. Extensive empirical studies demonstrate the effectiveness of the proposed framework.

Weakly supervised phrase grounding tasks aim to learn alignments between phrases and regions with coarse image-caption match information. One branch of previous methods established pseudo-label relationships between phrases and regions based on the Expectation-Maximization (EM) algorithm combined with contrastive learning. However, adopting a simplified batch-level local update (partial) of pseudo-labels in E-step is sub-optimal, while extending it to global update requires inefficiently numerous computations. In addition, their failure to consider potential false negative examples in contrastive loss negatively impacts the effectiveness of M-step optimization. To address these issues, we propose a Momentum Pseudo Labeling (MPL) method, which efficiently uses a momentum model to synchronize global pseudo-label updates on the fly with model parameter updating. Additionally, we explore potential relationships between phrases and regions from non-matching image-caption pairs and convert these false negative examples to positive ones in contrastive learning. Our approach achieved SOTA performance on 3 commonly used grounding datasets for weakly supervised phrase grounding tasks.

Word order difference between source and target languages is a major obstacle to cross-lingual transfer, especially in the dependency parsing task. Current works are mostly based on order-agnostic models or word reordering to mitigate this problem. However, such methods either do not leverage grammatical information naturally contained in word order or are computationally expensive as the permutation space grows exponentially with the sentence length. Moreover, the reordered source sentence with an unnatural word order may be a form of noising that harms the model learning. To this end, we propose an Implicit Word Reordering framework with Knowledge Distillation (IWR-KD). This framework is inspired by that deep networks are good at learning feature linearization corresponding to meaningful data transformation, e.g. word reordering. To realize this idea, we introduce a knowledge distillation framework composed of a word-reordering teacher model and a dependency parsing student model. We verify our proposed method on Universal Dependency Treebanks across 31 different languages and show it outperforms a series of competitors, together with experimental analysis to illustrate how our method works towards training a robust parser.

Text embeddings from large language models (LLMs) have achieved excellent results in tasks such as information retrieval, semantic textual similarity, etc. In this work, we show an interesting finding: when feeding a text into the LLM-based embedder, the obtained text embedding will be able to be aligned with the key tokens in the input text. We first fully analyze this phenomenon on eight LLM-based embedders and show that this phenomenon is universal and is not affected by model architecture, training strategy, and embedding method. With a deeper analysis, we find that the main change in embedding space between these embedders and their LLM backbones is in the first principal component. By adjusting the first principal component, we can align text embedding with the key tokens. Finally, we give several examples to demonstrate the vast application potential of this finding: (1) we propose a simple and practical sparse retrieval method based on the aligned tokens, which can achieve 80% of the dense retrieval effect of the same model while reducing the computation significantly; (2) we show that our findings provide a novel perspective to help understand novel technologies (e.g., instruction-following embedding) and fuzzy concepts (e.g., semantic relatedness vs. similarity) in this field.

Semi-supervised text classification (SSTC) aims to train text classification models with few labeled data and massive unlabeled data.Existing studies develop effective pseudo-labeling methods, but they can struggle with unlabeled data that have imbalanced classes mismatched with the labeled data, making the pseudolabeling biased towards majority classes, resulting in catastrophic error propagation.We believe it is crucial to explicitly estimate the overall class distribution, and use it to calibrate pseudo-labeling to constrain majority classes.To this end, we formulate the pseudo-labeling as an optimal transport (OT) problem, which transports the unlabeled sample distribution to the class distribution.With a memory bank, we dynamically collect both the high-confidence pseudo-labeled data and true labeled data, thus deriving reliable (pseudo-) labels for class distribution estimation.Empirical results on 3 commonly used benchmarks demonstrate that our model is effective and outperforms previous state-of-the-art methods.

Inductive logic programming (ILP) is a machine learning approach aiming to learn explanatory rules from data. While existing ILP systems can successfully solve small-scale tasks, large-scale applications with various language biases are rarely explored. Besides, it is crucial for a large majority of current ILP systems to require expert-defined language bias, which hampers the development of ILP towards broader utilizations. In this paper, we introduce GeoILP, a large-scale synthetic dataset of diverse ILP tasks involving numerous aspects of language bias. These tasks are built from geometry problems, at the level from textbook exercise to regional International Mathematical Olympiad (IMO), with the help of a deduction engine. These problems are elaborately selected to cover all challenging language biases, such as recursion, predicate invention, and high arity. Experimental results show that no existing method can solve GeoILP tasks. In addition, along with classic symbolic-form data, we provide image-form data to boost the development of the joint learning of neural perception and symbolic rule induction.

Homomorphism is an important structure-preserving mapping between graphs. Given a graph G and a pattern Q, the subgraph homomorphism problem is to find a mapping φ from Q to G such that adjacent vertices of Q are mapped to adjacent vertices in G. Unlike the subgraph isomorphic mapping that is injective, homomorphism allows multiple vertices in Q to map to the same vertex in G, increasing complexity. We develop HFrame, the first GNN-based framework for subgraph homomorphism, by combining algorithms and machine learning. We show that HFrame is more expressive than the vanilla GNN, i.e., HFrame can distinguish more graph pairs (Q, G) such that Q is not homomorphic to G. Moreover, we provide a generalization error bound for HFrame. Using real-life and synthetic graphs, we show that HFrame is up to 101.91× faster than exact matching algorithms, and its average accuracy can reach 0.962.

Hierarchical text classification (HTC) aims to assign one or more labels in the hierarchy for each text. Many methods represent this structure as a global hierarchy, leading to redundant graph structures. To address this, incorporating a text-specific local hierarchy is essential. However, existing approaches often model this local hierarchy as a sequence, focusing on explicit parent-child relationships while ignoring implicit correlations among sibling/peer relationships. In this paper, we first integrate local hierarchies into a manual depth-level prompt to capture parent-child relationships. We then apply Mixup to this hierarchical prompt tuning scheme to improve the latent correlation within sibling/peer relationships. Notably, we propose a novel Mixup ratio guided by local hierarchy correlation to effectively capture intrinsic correlations. This Local Hierarchy Mixup (LH-Mix) model demonstrates remarkable performance across three widely-used datasets.

Recent Text-to-SQL methods leverage large language models (LLMs) by incorporating feedback from the database management system. While these methods effectively address execution errors in SQL queries, they struggle with database mismatches--errors that do not trigger execution exceptions. Database mismatches include issues such as condition mismatches and stricter constraint mismatches, both of which are more prevalent in real-world scenarios. To address these challenges, we propose a tool-assisted agent framework for SQL verification and refinement, equipping the LLM-based agent with two specialized tools: a retriever and a detector, designed to diagnose and correct SQL queries with database mismatches. These tools enhance the capability of LLMs to handle real-world questions more effectively. We also introduce SpiderMismatch, a new dataset specifically constructed to reflect the condition mismatch problems encountered in real-world scenarios. Empirical studies demonstrate the effectiveness of our proposed model on Spider and Spider-Realistic datasets in few-shot settings and confirm that our model outperforms baseline methods on SpiderMismatch.

Xiyang Liu, Chunming Hu, Richong Zhang, Junfan Chen, Baowen Xu. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

Inductive reasoning infers general rules from observed evidence, which is one of the most critical intelligence abilities. Previous works have succeeded in formal languages but suffer from onerous and error-prone conversions between a particular formal language and the working language. As large language models (LLMs) have emerged, direct reasoning with various kinds of languages, especially natural languages, without formal language involvement has become feasible. However, existing LLM-based inductive reasoning usually relies on LLM's intrinsic generation ability, which is prone to LLM's hallucination and lacks systematic guidance according to the nature of inductive reasoning. To this end, we propose HypoBootstrap, an integrated framework for inductive reasoning that generates and confirms hypotheses both in a bootstrapping manner. Regarding hypothesis generation, we propose a novel bootstrapping generation strategy, bootstrapping object hypotheses, relational hypotheses, and functional hypotheses successively, which assists LLM in observing the evidence from trivial patterns to non-trivial patterns. Regarding hypothesis confirmation, we utilize Glymour's theory of bootstrap confirmation, a hypothesis confirmation theory from the philosophy of science that can confirm a set of hypotheses. We use its principles to confirm the object hypotheses, relational hypotheses, and functional hypotheses. Empirical studies on four inductive reasoning scenarios of different natures, involving causal induction, concept learning, grammar learning, and abstract reasoning, demonstrate that HypoBootstrap significantly outperforms existing methods.

Multi-label text classification (MLTC) assigns multiple labels to a sentence, with the key challenge being capturing label correlations. Existing models prioritize leveraging correlations but often overlook overfitting, while plug-and-play regularization methods fail to preserve correlations effectively. In this paper, we distinguish two types of label correlations: explicit co-occurring correlations and implicit semantic correlations, and propose regularizations on prototypical label embeddings for correlation preservation. Specifically, we first generate the prototypical embedding of multiple co-occurred labels as an intermediate. We then apply a prototypical regularization on the distance between the sentence embedding and corresponding prototypical embedding to alleviate the over-alignment issue caused by binary cross entropy loss and facilitate explicit correlation preservation. We finally extend the vanilla Mixup, which solely mixes multi-hot labels, on prototypical embedding mixing to promote implicit correlation preservation. Empirical studies show the effectiveness of our regularization methods.

Knowledge Graph Completion (KGC) aims to complete KGs by predicting missing entities. A common solution for KGC is Knowledge Graph Embedding (KGE), which assumes that semantical similar entities or relationships should possess similar representations in high-dimensional space. In KGE, a heuristic score function of the head entity and its relation with different operators is required. A typical technique is regularization for tensor factorization, such as the Nuclear-p norm and the Frobenius norm of the query/entity embedding, which significantly improve the KGE model performance on the KGC task. However, the Co-Relation s, including the association between tail entities ( Co-Query Relation ) and the association between queries ( Co-Entity Relation ), desirable for KGC are not fully considered in existing embedding regularization techniques. In this article, we theoretically interpret the role of Co-Relation in KGE and propose a novel ConR regularization approach to learn embedding that takes Co-Relations into account. Extensive experiments show that our model improves static and temporal KGC tasks over decomposition-based models, ComplEx and TuckER. Further analysis of the score cumulative distribution function and embedding visualization demonstrates the effectiveness of ConR .

To date, a backbone of methods for unsupervised domain adaptation (UDA) involves learning label-discriminative features via a label classifier and domain-invariant features through a domain discriminator in an adversarial scheme. However, these methods lack explicit control for aligning the source data and target data within the same label class, degrading the classifier's performance in the target domain. In this paper, we propose PL-Mix, a pseudo label guided Mixup method based on adversarial prompt tuning. Specifically, our PL-Mix facilitates class-dependent alignment and can alleviate the impact of noisy pseudo-labels. We then theoretically justify that PL-Mix can improve the generalization for UDA. Extensive experiments of the comparison with existing models also demonstrate the effectiveness of PL-Mix.

Sentence Representation Learning (SRL) is a fundamental task in Natural Language Processing (NLP), with the Contrastive Learning of Sentence Embeddings (CSE) being the mainstream technique due to its superior performance. An intriguing phenomenon in CSE is the significant performance gap between supervised and unsupervised methods, with their only difference lying in the training data. Previous works attribute this performance gap to differences in two representation properties (alignment and uniformity). However, since alignment and uniformity only measure the results, they fail to answer "What aspects of the training data contribute to the performance gap?" and "How can the performance gap be narrowed?", In this paper, we conduct empirical experiments to answer these "What" and "How" questions. We first answer the "What" question by thoroughly comparing the behavior of supervised and unsupervised CSE during their respective training processes. From the comparison, we identify the similarity pattern as a key factor to the performance gap, and introduce a metric, called Relative Fitting Difficulty (RFD), to measure the complexity of the similarity pattern. Then, based on the insights gained from the "What" question, we tackle the "How" question by increasing the pattern complexity of the training data. We achieve this by leveraging the In-Context Learning (ICL) capability of the Large Language Model (LLM) to generate data that simulates complex patterns. By utilizing the hierarchical patterns in the LLM-generated data, we effectively narrow the gap between supervised and unsupervised CSE. We release our codes and appendix at https://github.com/BDBC-KG-NLP/NGCSE.

Current methods for Knowledge-Based Question Answering (KBQA) usually rely on complex training techniques and model frameworks, leading to many limitations in practical applications. Recently, the emergence of In-Context Learning (ICL) capabilities in Large Language Models (LLMs) provides a simple and training-free semantic parsing paradigm for KBQA: Given a small number of questions and their labeled logical forms as demo examples, LLMs can understand the task intent and generate the logic form for a new question. However, current powerful LLMs have little exposure to logic forms during pre-training, resulting in a high format error rate. To solve this problem, we propose a code-style in-context learning method for KBQA, which converts the generation process of unfamiliar logical form into the more familiar code generation process for LLMs. Experimental results on three mainstream datasets show that our method dramatically mitigated the formatting error problem in generating logic forms while realizing a new SOTA on WebQSP, GrailQA, and GraphQ under the few-shot setting. The code and supplementary files are released at https://github.com/Arthurizijar/KB-Coder.

Open-Set Semi-Supervised Text Classification (OSTC) aims to train a classification model on a limited set of labeled texts, alongside plenty of unlabeled texts that include both in-distribution and out-of-distribution examples.In this paper, we revisit the main challenge in OSTC, i.e., outlier detection, from a measurement disagreement perspective and innovatively propose to improve OSTC performance by directly maximizing the measurement disagreements.Based on the properties of in-measurement and crossmeasurements, we design an Adversarial Disagreement Maximization (ADM) model that synergeticly optimizes the measurement disagreements.In addition, we develop an abnormal example detection and measurement calibration approach to guarantee the effectiveness of ADM training.Experiment results and comprehensive analysis of three benchmarks demonstrate the effectiveness of our model.

Multi-Modal Entity Alignment aims to discover identical entities across heterogeneous knowledge graphs.While recent studies have delved into fusion paradigms to represent entities holistically, the elimination of features irrelevant to alignment and modal inconsistencies is overlooked, which are caused by inherent differences in multi-modal features.To address these challenges, we propose a novel strategy of progressive modality freezing, called PMF, that focuses on alignmentrelevant features and enhances multi-modal feature fusion.Notably, our approach introduces a pioneering cross-modal association loss to foster modal consistency.Empirical evaluations across nine datasets confirm PMF's superiority, demonstrating stateof-the-art performance and the rationale for freezing modalities.

Sentence Representation Learning (SRL) is a crucial task in Natural Language Processing (NLP), where contrastive Self-Supervised Learning (SSL) is currently a mainstream approach. However, the reasons behind its remarkable effectiveness remain unclear. Specifically, many studies have investigated the similarities between contrastive and non-contrastive SSL from a theoretical perspective. Such similarities can be verified in classification tasks, where the two approaches achieve comparable performance. But in ranking tasks (i.e., Semantic Textual Similarity (STS) in SRL), contrastive SSL significantly outperforms non-contrastive SSL. Therefore, two questions arise: First, what commonalities enable various contrastive losses to achieve superior performance in STS? Second, how can we make non-contrastive SSL also effective in STS? To address these questions, we start from the perspective of gradients and discover that four effective contrastive losses can be integrated into a unified paradigm, which depends on three components: the Gradient Dissipation, the Weight, and the Ratio. Then, we conduct an in-depth analysis of the roles these components play in optimization and experimentally demonstrate their significance for model performance. Finally, by adjusting these components, we enable non-contrastive SSL to achieve outstanding performance in STS.

Temporal Knowledge Graph Forecasting (TKGF) aims to forecast the missing entities or relations at a specific timestamp when only the historical information is observed. It is crucial to accurately identify the historical information of complex temporal relational graphs related to the query. Existing works, e.g., TANGO, have exploited the Neural Ordinary Differential Equation (NODE) to TKGF. However, TANGO encounters two limitations. First, TANGO observes historical facts with only one timestamp at each step, leading to a long-term forgetting problem. Second, TANGO gives the same weight to the entire history graph, including facts that are not relevant to the query. To tackle the above limitations, this paper utilizes Attentional Neural Integral Equation for TKGF (tIE), enabling the global interaction between query-related historical graph sequences. To achieve this, we employ the Relational Graph Convolutional Network and Fourier-type Transformer to model the graph structure and temporal evolution of TKG. The Iterative Integral Equation Solver is exploited to enhance the accuracy and robustness of numerical solutions. The proposed method outperforms baseline models regarding several metrics and inference speed on four benchmark datasets, especially on the long horizontal link forecasting task with irregular time intervals.

Cross-lingual Cross-modal Retrieval (CCR) is an essential task in web search, which aims to break the barriers between modality and language simultaneously and achieves image-text retrieval in the multi-lingual scenario with a single model. In recent years, excellent progress has been made based on cross-lingual cross-modal pre-training; particularly, the methods based on contrastive learning on large-scale data have significantly improved retrieval tasks. However, these methods directly follow the existing pre-training methods in the cross-lingual or cross-modal domain, leading to two problems of inconsistency in CCR: The methods with cross-lingual style suffer from the intra-modal error propagation, resulting in inconsistent recall performance across languages in the whole dataset. The methods with cross-modal style suffer from the inter-modal optimization direction bias, resulting in inconsistent rank across languages within each instance, which cannot be reflected by Recall@K. To solve these problems, we propose a simple but effective 1-to-K contrastive learning method, which treats each language equally and eliminates error propagation and optimization bias. In addition, we propose a new evaluation metric, Mean Rank Variance (MRV), to reflect the rank inconsistency across languages within each instance. Extensive experiments on four CCR datasets show that our method improves both recall rates and MRV with smaller-scale pre-trained data, achieving the new state-of-art.

The Composed Image Retrieval (CIR) task aims to retrieve target images using a composed query consisting of a reference image and a modified text. Advanced methods often utilize contrastive learning as the optimization objective, which benefits from adequate positive and negative examples. However, the triplet for CIR incurs high manual annotation costs, resulting in limited positive examples. Furthermore, existing methods commonly use in-batch negative sampling, which reduces the negative number available for the model. To address the problem of lack of positives, we propose a data generation method by leveraging a multi-modal large language model to construct triplets for CIR. To introduce more negatives during fine-tuning, we design a two-stage fine-tuning framework for CIR, whose second stage introduces plenty of static representations of negatives to optimize the representation space rapidly. The above two improvements can be effectively stacked and designed to be plug-and-play, easily applied to existing CIR models without changing their original architectures. Extensive experiments and ablation analysis demonstrate that our method effectively scales positives and negatives and achieves state-of-the-art results on both FashionIQ and CIRR datasets. In addition, our method also performs well in zero-shot composed image retrieval, providing a new CIR solution for the low-resources scenario. Our code and data are released at https://github.com/BUAADreamer/SPN4CIR.

Text classification is one vital tool assisting web content mining. Semi-supervised text classification (SSTC) offers an approach to alleviate the burden of annotation costs by training on a few labeled texts alongside many unlabeled texts. Unsolved challenges in SSTC are the overfitting problem caused by the limited labeled data and the mislabeling problem of unlabeled texts. To address these issues, this paper proposes a Self-Paced PairWise representation learning (SPPW) model. Concretely, SPPW alleviates the overfitting problem by replacing the overfitting-prone learning of a parameterized classifier with representation learning in a pair-wise manner. Besides, we propose a novel self-paced text filtering method that effectively integrates both label confidence and text hardness to reduce mislabeled texts synergistically. Extensive experiments on 3 benchmark SSTC datasets show that SPPW outperforms baselines and is effective in mitigating overfitting and mislabeling problems.

Text classification is a fundamental task in web content mining. Although the existing supervised contrastive learning (SCL) approach combined with pre-trained language models (PLMs) has achieved leading performance in text classification, it lacks fundamental principles. Theoretically motivated by a derived lower bound of mutual information maximization, we propose a dual contrastive learning framework DualCL that satisfies three properties, i.e., parameter-free, augmentation-easy and label-aware. DualCL generates classifier parameters from the PLM and simultaneously uses them for classification and as augmented views of the input text for supervised contrastive learning. Extensive experiments conclusively demonstrate that DualCL excels in learning superior text representations and consistently outperforms baseline models.

Multimodal relation extraction is a fundamental task of multimodal information extraction. Recent studies have shown promising results by integrating hierarchical visual features from local regions, like image patches, to the broader global regions that form the entire image. However, research to date has largely ignored the understanding of how hierarchical visual semantics are represented and the characteristics that can benefit relation extraction. To bridge this gap, we propose a novel two-stage hierarchical visual context fusion transformer incorporating the mixture of multimodal experts framework to effectively represent and integrate hierarchical visual features into textual semantic representations. In addition, we introduce the concept of hierarchical tracking maps to facilitate the understanding of the intrinsic mechanisms of image information processing involved in multimodal models. We thoroughly investigate the implications of hierarchical visual contexts through four dimensions: performance evaluation, the nature of auxiliary visual information, the patterns observed in the image encoding hierarchy, and the significance of various visual encoding levels. Empirical studies show that our approach achieves new state-of-the-art performance on the MNRE dataset.

Author name disambiguation (AND) is an essential task for online academic retrieval systems. Recent models adopt representation learning in the author's name disambiguation. Despite achieving remarkable success, these methods may be limited in two aspects. First, the heuristically constructed paper association graphs used for representation learning contain uncertainties that may cause negative supervision. Second, existing algorithms, such as binary cross-entropy loss, used to train representation learning models may not produce sufficiently high-quality representations for AND. To tackle the above problems, we propose an association refining and compositional contrasting (ARCC) framework for AND tasks. ARCC first adopts an iterative graph structure refinement process to dynamically reduce the uncertainties in paper graphs. Then, a compositional contrastive learning method is proposed to encourage learning more discriminative representations for AND. Empirical studies on two benchmark datasets suggest that ARCC is effective for AND and outperforms the state-of-the-art models.

Cross-lingual learning aims to transfer knowledge from one natural language to another. Zero-shot cross-lingual named entity recognition (NER) tasks are to train an NER model on source languages and to identify named entities in other languages. Existing knowledge distillation-based models in a teacher-student manner leverage the unlabeled samples from the target languages and show their superiority in this setting. However, the valuable similarity information between tokens in the target language is ignored. And the teacher model trained solely on the source language generates low-quality pseudo-labels. These two facts impact the performance of cross-lingual NER. To improve the reliability of the teacher model, in this study, we first introduce one extra simple binary classification teacher model by similarity learning to measure if the inputs are from the same class. We note that this binary classification auxiliary task is easier, and the two teachers simultaneously supervise the student model for better performance. Furthermore, given such a stronger student model, we propose a progressive knowledge distillation framework that extensively fine-tunes the teacher model on the target-language pseudo-labels generated by the student model. Empirical studies on three datasets across seven different languages show that our presented model outperforms state-of-the-art methods.

Meta-learning has recently promoted few-shot text classification, which identifies target classes based on information transferred from source classes through a series of small tasks or episodes. Existing works constructing their meta-learner on Prototypical Networks need improvement in learning discriminative text representations between similar classes that may lead to conflicts in label prediction. The overfitting problems caused by a few training instances need to be adequately addressed. In addition, efficient episode sampling procedures that could enhance few-shot training should be utilized. To address the problems mentioned above, we first present a contrastive learning framework that simultaneously learns discriminative text representations via supervised contrastive learning while mitigating the overfitting problem via unsupervised contrastive regularization, and then we build an efficient self-paced episode sampling approach on top of it to include more difficult episodes as training progresses. Empirical results on eight few-shot text classification datasets show that our model outperforms the current state-of-the-art models. The extensive experimental analysis demonstrates that our supervised contrastive representation learning and unsupervised contrastive regularization techniques improve the performance of few-shot text classification. The episode-sampling analysis reveals that our self-paced sampling strategy improves training efficiency.

Data augmentation is widely used in text classification, especially in the low-resource regime where a few examples for each class are available during training. Despite the success, generating data augmentations as hard positive examples that may increase their effectiveness is under-explored. This paper proposes an Adversarial Word Dilution (AWD) method that can generate hard positive examples as text data augmentations to train the low-resource text classification model efficiently. Our idea of augmenting the text data is to dilute the embedding of strong positive words by weighted mixing with unknown-word embedding, making the augmented inputs hard to be recognized as positive by the classification model. We adversarially learn the dilution weights through a constrained min-max optimization process with the guidance of the labels. Empirical studies on three benchmark datasets show that AWD can generate more effective data augmentations and outperform the state-of-the-art text data augmentation methods. The additional analysis demonstrates that the data augmentations generated by AWD are interpretable and can flexibly extend to new examples without further training.

Prompt-based Learning has shown significant success in few-shot classification. The mainstream approach is to concatenate a template for the input text to transform the classification task into a cloze-type task where label mapping plays an important role in finding the ground-truth labels. While current label mapping methods only use the contexts in one single input, it could be crucial if wrong information is contained in the text. Specifically, it is proved in recent work that even the large language models like BERT/RoBERTa make classification decisions heavily dependent on a specific keyword regardless of the task or the context. Such a word is referred to as a lexical cue and if a misleading lexical cue is included in the instance it will lead the model to make a wrong prediction. We propose a multi-mask prompt-based approach with Multi-Mask Label Mapping (MMLM) to reduce the impact of misleading lexical cues by allowing the model to exploit multiple lexical cues. To satisfy the conditions of few-shot learning, an instance augmentation approach for the cloze-type model is proposed and the misleading cues are gradually excluded through training. We demonstrate the effectiveness of MMLM by both theoretical analysis and empirical studies, and show that MMLM outperforms other existing label mapping approaches.

The semi-supervised text classification (SSTC) task aims at training text classification models with a few labeled data and massive unlabeled data.Recent works achieve this task by pseudo-labeling methods that assign pseudolabels to unlabeled data as additional supervision.However, these models may suffer from incorrect pseudo-labels caused by underfitting of decision boundaries and generating biased pseudo-labels on imbalanced data.We propose a prototype-guided semi-supervised model to address the above problems, which integrates a prototype-anchored contrasting strategy and a prototype-guided pseudo-labeling strategy.Particularly, the prototype-anchored constrasting constructs prototypes to cluster text representations with the same class, forcing them to be high-density distributed, thus alleviating the underfitting of decision boundaries.And the prototype-guided pseudo-labeling selects reliable pseudo-labeled data around prototypes based on data distribution, thus alleviating the bias from imbalanced data.Empirical results on 4 commonly-used datasets demonstrate that our model is effective and outperforms state-ofthe-art methods.

Document-level relation extraction (DocRE)involves identifying relations between entities distributed in multiple sentences within a document.Existing methods focus on building a heterogeneous document graph to model the internal structure of an entity and the external interaction between entities.However, there are two drawbacks in existing methods.On one hand, anaphor plays an important role in reasoning to identify relations between entities but is ignored by these methods.On the other hand, these methods achieve crosssentence entity interactions implicitly by utilizing a document or sentences as intermediate nodes.Such an approach has difficulties in learning fine-grained interactions between entities across different sentences, resulting in sub-optimal performance.To address these issues, we propose an Anaphor-Assisted (AA) framework for DocRE tasks.Experimental results on the widely-used datasets demonstrate that our model achieves a new state-of-the-art performance.1

Semi-supervised text classification (STC) has been extensively researched and reduces human annotation. However, existing research assuming that unlabeled data only contains in-distribution texts is unrealistic. This paper extends STC to a more practical Open-set Semi-supervised Text Classification (OSTC) setting, which assumes that the unlabeled data contains out-of-distribution (OOD) texts. The main challenge in OSTC is the false positive inference problem caused by inadvertently including OOD texts during training. To address the problem, we first develop baseline models using outlier detectors for hard OOD-data filtering in a pipeline procedure. Furthermore, we propose a Latent Outlier Softening (LOS) framework that integrates semi-supervised training and outlier detection within probabilistic latent variable modeling. LOS softens the OOD impacts by the Expectation-Maximization (EM) algorithm and weighted entropy maximization. Experiments on 3 created datasets show that LOS significantly outperforms baselines.

Opinion target extraction (OTE) or aspect extraction (AE) is a fundamental task in opinion mining that aims to extract the targets (or aspects) on which opinions have been expressed. Recent work focus on cross-domain OTE, which is typically encountered in real-world scenarios, where the testing and training distributions differ. Most methods use domain adversarial neural networks that aim to reduce the domain gap between the labelled source and unlabelled target domains to improve target domain performance. However, this approach only aligns feature distributions and does not account for class-wise feature alignment, leading to suboptimal results. Semi-supervised learning (SSL) has been explored as a solution, but is limited by the quality of pseudo-labels generated by the model. Inspired by the theoretical foundations in domain adaptation [2], we propose a new SSL approach that opts for selecting target samples whose model output from a domain-specific teacher and student network disagree on the unlabelled target data, in an effort to boost the target domain performance. Extensive experiments on benchmark cross-domain OTE datasets show that this approach is effective and performs consistently well in settings with large domain shifts.

Word Sense Disambiguation (WSD) which aims to identify the correct sense of a target word appearing in a specific context is essential for web text analysis. The use of glosses has been explored as a means for WSD. However, only a few works model the correlation between the target context and gloss. We add to the body of literature by presenting a model that employs a multi-head attention mechanism on deep contextual features of the target word and candidate glosses to refine the target word embedding. Furthermore, to encourage the model to learn the relevant part of target features that align with the correct gloss, we recursively alternate attention on target word features and that of candidate glosses to gradually extract the relevant contextual features of the target word, refining its representation and strengthening the final disambiguation results. Empirical studies on the five most commonly used benchmark datasets show that our proposed model is effective and achieves state-of-the-art results.

We consider the problem of learning a neural network classifier. Under the information bottleneck (IB) principle, we associate with this classification problem a representation learning problem, which we call "IB learning". We show that IB learning is, in fact, equivalent to a special class of the quantization problem. The classical results in rate-distortion theory then suggest that IB learning can benefit from a "vector quantization" approach, namely, simultaneously learning the representations of multiple input objects. Such an approach assisted with some variational techniques, result in a novel learning framework, "Aggregated Learning", for classification with neural network models. In this framework, several objects are jointly classified by a single neural network. The effectiveness of this framework is verified through extensive experiments on standard image recognition and text classification tasks.

Multi-label text classification (MLTC) has a wide range of real-world applications. Neural networks recently promoted the performance of MLTC models. Training these neural-network models relies on sufficient accurately labelled data. However, manually annotating large-scale multi-label text classification datasets is expensive and impractical for many applications. Weak supervision techniques have thus been developed to reduce the cost of annotating text corpus. However, these techniques introduce noisy labels into the training data and may degrade the model performance. This paper aims to deal with such noise-label problems in MLTC in both single-instance and multi-instance settings. We build a novel Neural Expectation-Maximization Framework (nEM) that combines neural networks with probabilistic modelling. The nEM framework produces text representations using neural-network text encoders and is optimized with the Expectation-Maximization algorithm. It naturally considers the noisy labels during learning by iteratively updating the model parameters and estimating the distribution of the ground-truth labels. We evaluate our nEM framework in multi-instance noisy MLTC on a benchmark relation extraction dataset constructed by distant supervision and in single-instance noisy MLTC on synthetic noisy datasets constructed by keywords supervision and label flipping. The experimental results demonstrate that nEM significantly improves upon baseline models in both single-instance and multi-instance noisy MLTC tasks. The experiment analysis suggests that our nEM framework efficiently reduces the noisy labels in MLTC datasets and significantly improves model performance.

Entity alignment is a vital task in knowledge fusion, which aims to align entities from different knowledge graphs and merge them into one single graph. Existing entity alignment models focus on local features and try to minimize the distance between pairs of pre-aligned entities. Despite their success, these models heavily rely on the number of existing pre-aligned entity pairs and the topology information from the rest large set of unaligned entities is still largely unexplored. To overcome the limitation of existing models, we propose a model, termed Global Alignment and Local Information Aggregation, or GALA. GALA constructs global features for the knowledge graphs to be aligned using entity embeddings. It aligns the entities in the graphs by forcing their global features to match with each other and progressively updating the entity embeddings by aggregating local information from the other network. Empirical studies on commonly-used KG alignment data sets confirm the effectiveness of the proposed model.

Information extraction aims at extracting entities, relations, and so on, in text to support information retrieval systems. To extract information, researchers have considered multitask learning (ML) approaches. The conventional ML approach learns shared features across tasks, with the assumption that these features capture sufficient task interactions to learn expressive shared representations for task classification. However, such an assumption is flawed in different perspectives. First, the shared representation may contain noise introduced by another task; tasks coupled for multitask learning may have different complexities but this approach treats all tasks equally; the conventional approach has a flat structure that hinders the learning of explicit interactions. This approach, however, learns implicit interactions across tasks and often has a generalization ability that has benefited the learning of multitasks. In this article, we take advantage of implicit interactions learned by conventional approaches while alleviating the issues mentioned above by developing a Recurrent Interaction Network with an effective Early Prediction Integration (RIN-EPI) for multitask learning. Specifically, RIN-EPI learns implicit and explicit interactions across two different but related tasks. To effectively learn explicit interactions across tasks, we consider the correlations among the outputs of related tasks. It is, however, obvious that task outputs are unobservable during training, so we leverage the predictions at intermediate layers (referred to as early predictions) as proxies as well as shared features across tasks to learn explicit interactions through attention mechanisms and sequence learning models. By recurrently learning explicit interactions, we gradually improve predictions for the individual tasks in the multitask learning. We demonstrate the effectiveness of RIN-EPI on the learning of two mainstream multitasks for information extraction: (1) entity recognition and relation classification and (2) aspect and opinion term co-extraction. Extensive experiments demonstrate the effectiveness of the RIN-EPI architecture, where we achieve state-of-the-art results on several benchmark datasets.

Each link prediction task requires different degrees of answer diversity. While a link prediction task may expect up to a couple of answers, another may expect nearly a hundred answers. Given this fact, the performance of a link prediction model can be estimated more accurately if a flexible number of obtained answers are estimated instead of a predefined number of answers. Inspired by this, in this article, we analyze two evaluation criteria for link prediction tasks, respectively ranking-based protocol and sampling-based protocol. Furthermore, we study two classes of models on link prediction task, direct model and latent-variable model respectively, to demonstrate that latent-variable model performs better under the sampling-based protocol. We then propose a latent-variable model where the framework of Conditional Variational AutoEncoder (CVAE) is applied. Experimental study suggests that the proposed model performs comparably to the current state-of-the-art even under the conventional rank-based protocol. Under the sampling-based protocol, the proposed model is shown to outperform various state-of-the-art models.

Few-shot text classification has recently been promoted by the meta-learning paradigm which aims to identify target classes with knowledge transferred from source classes with sets of small tasks named episodes. Despite their success, existing works building their meta-learner based on Prototypical Networks are unsatisfactory in learning discriminative text representations between similar classes, which may lead to contradictions during label prediction. In addition, the task-level and instance-level overfitting problems in few-shot text classification caused by a few training examples are not sufficiently tackled. In this work, we propose a contrastive learning framework named ContrastNet to tackle both discriminative representation and overfitting problems in few-shot text classification. ContrastNet learns to pull closer text representations belonging to the same class and push away text representations belonging to different classes, while simultaneously introducing unsupervised contrastive regularization at both task-level and instance-level to prevent overfitting. Experiments on 8 few-shot text classification datasets show that ContrastNet outperforms the current state-of-the-art models.

Unsupervised sentence representation learning is a fundamental problem in natural language processing. Recently, contrastive learning has made great success on this task. Existing constrastive learning based models usually apply random sampling to select negative examples for training. Previous work in computer vision has shown that hard negative examples help contrastive learning to achieve faster convergency and better optimization for representation learning. However, the importance of hard negatives in contrastive learning for sentence representation is yet to be explored. In this study, we prove that hard negatives are essential for maintaining strong gradient signals in the training process while random sampling negative examples is ineffective for sentence representation. Accordingly, we present a contrastive model, MixCSE, that extends the current state-of-the-art SimCSE by continually constructing hard negatives via mixing both positive and negative features. The superior performance of the proposed approach is demonstrated via empirical studies on Semantic Textual Similarity datasets and Transfer task datasets.

Cross-lingual named entity recognition task is one of the critical problems for evaluating the potential transfer learning techniques on low resource languages.Knowledge distillation using pre-trained multilingual language models between source and target languages have shown their superiority in transfer.However, existing cross-lingual distillation models merely consider the potential transferability between two identical single tasks across both domains.Other possible auxiliary tasks to improve the learning performance have not been fully investigated.In this study, based on the knowledge distillation framework and multitask learning, we introduce the similarity metric model as an auxiliary task to improve the cross-lingual NER performance on the target domain.Specifically, an entity recognizer and a similarity evaluator are first trained in parallel as two teachers from the source domain.Then, two tasks in the student model are supervised by these teachers simultaneously.Empirical studies on the three datasets across 7 different languages confirm the effectiveness of the proposed model.

Constructing a robust conversation graph based false information detection model is crucial for real social platforms. Recently, graph neural network (GNN) methods for false information detection have achieved significant advances. However, we empirically find that slight perturbations in the conversation graph can cause the predictions of existing models to collapse. To address this problem, we present RDCL, a contrastive learning framework for false information detection on social networks, to obtain robust detection results. RDCL leverages contrastive learning to maximize the consistency between perturbed graphs from the same original graph and minimize the distance between perturbed and original graphs from the same class, forcing the model to improve resistance to data perturbations. Moreover, we prove the importance of hard positive samples for contrastive learning and propose a hard positive sample pairs generation method (HPG) for conversation graphs, which can generate stronger gradient signals to improve the contrastive learning effect and make the model more robust. Experiments on various GNN encoders and datasets show that RDCL outperforms the current state-of-the-art models.

Enhancing the interpretability of text classification models can help increase the reliability of these models in real-world applications. Currently, most researchers focus on extracting task-specific words from inputs to improve the interpretability of the model. The competitive approaches exploit the Variational Information Bottleneck (VIB) to improve the performance of word masking at the word embedding layer to obtain task-specific words. However, these approaches ignore the multi-level semantics of the text, which can impair the interpretability of the model, and do not consider the risk of representation overlap caused by the VIB, which can impair the classification performance. In this paper, we propose an enhanced variational word masks approach, named E-VarM, to solve these two issues effectively. The E-VarM combines multi-level semantics from all hidden layers of the model to mask out task-irrelevant words and uses contrastive learning to readjust the distances between representations. Empirical studies on ten benchmark text classification datasets demonstrate that our approach outperforms the SOTA methods in simultaneously improving the interpretability and accuracy of the model.

Weakly supervised phrase grounding aims to learn an alignment between phrases in a caption and objects in a corresponding image using only caption-image annotations, i.e., without phrase-object annotations.Previous methods typically use a caption-image contrastive loss to indirectly supervise the alignment between phrases and objects, which hinders the maximum use of the intrinsic structure of the multimodal data and leads to unsatisfactory performance.In this work, we directly use the phrase-object contrastive loss in the condition that no positive annotation is available in the first place.Specifically, we propose a novel contrastive learning framework based on the expectation-maximization algorithm that adaptively refines the target prediction.Experiments on two widely used benchmarks, Flickr30K Entities and RefCOCO+, demonstrate the effectiveness of our framework.We obtain 63.05% top-1 accuracy on Flickr30K Entities and 59.51%/43.46% on RefCOCO+ TestA/TestB, outperforming the previous methods by a large margin, even surpassing a previous SoTA that uses a pre-trained visionlanguage model.Furthermore, we deliver a theoretical analysis of the effectiveness of our method from the perspective of the maximum likelihood estimate with latent variables.

Overfitting is a common problem when there is insufficient data to train deep neural networks in machine learning tasks.Data augmentation regularization methods such as Dropout, Mixup, and their enhanced variants, are effective and prevalent, and achieve promising performance to overcome overfitting.However, in text learning, most of the existing regularization approaches merely adopt ideas from computer vision without considering the importance of dimensionality in natural language processing.In this paper, we argue that the property is essential to overcome overfitting in text learning.Accordingly, we present a saliency map informed textual data augmentation and regularization framework, which combines Dropout and Mixup, namely DropMix, to mitigate the overfitting problem in text learning.In addition, we design a procedure that drops and patches fine grained shapes of the saliency map under the DropMix framework to enhance regularization.Empirical studies confirm the effectiveness of the proposed approach on 12 text classification tasks.

Text style transfer is an important task in natural language processing with broad applications.Existing models following the masking and filling scheme suffer two challenges: the word masking procedure may mistakenly remove unexpected words and the selected words in the word filling procedure may lack diversity and semantic consistency.To tackle both challenges, in this study, we propose a style transfer model, with an adversarial masking approach and a styled filling technique (AMSF).Specifically, AMSF first trains a mask predictor by adversarial training without manual configuration.Then two additional losses, i.e. an entropy maximization loss and a consistency regularization loss, are introduced in training the word filling module to guarantee the diversity and semantic consistency of the transferred texts.Experimental results and analysis on two benchmark text style transfer data sets demonstrate the effectiveness of the proposed approaches.

This paper considers the domain generalization problem in which the labels of the query batch may be predicted at once and the model is allowed to assign a label to a query example utilizing all information contained in the query batch. In this setting, we identify two problems in the standard adaptive risk minimization (ARM) framework, namely, mismatched context and overfitted context, which are most severely manifested respectively at small and large support batch sizes. The existence of these problems signifies a fundamental tradeoff between within-domain learning and cross-domain generalization. We also propose two context adversarial training approaches to alleviate these problems so as to achieve a better tradeoff. We demonstrate experimentally that the proposed approaches uniformly outperform the standard ARM training for all choices of support batch sizes.

Fine-Grained entity typing is complicated by the fact that type labels form a hierarchical structure, and those training examples usually contain noisy type labels. This paper addresses these two issues by proposing a novel framework that simultaneously models the correlation among hierarchical types and the noise within the training data. Additionally, the framework contains an innovative training approach during which the noise in the training data is progressively removed. Experiments on standard benchmarking datasets validate the proposed framework and establish it as a new state of the art for this problem.

Aspect-level sentiment classification (ALSC); a fine-grained task of sentiment analysis holds the promise of machines communicating knowledge of different aspects of a product/service with humans. Specifically, ALSC aims at inferring the sentiment expressed toward a specific aspect term in a user-generated textual content. It is a long-standing problem in sentiment analysis, and its usefulness cannot be underestimated. Recent studies employ the dependency tree and restructure it to model the syntactic relationship among words in text. However, the restructuring process destructs the structural information of the original dependency tree. As a solution, we first construct a syntax graph that preserves the structural information of the dependency tree. We then propose a reliable syntax-based neural network model that performs a thorough search on the syntax graph to effectively find the relevant contextual information with respect to the aspect term for the sentence encoding. Noting that dependency trees parsed from existing dependency parsers may contain incorrect syntactic dependencies due to grammatical errors in a sentence, we adopt a convolutional layer that takes into account the relations among features of words in a local neighborhood to mitigate the issues brought by incorrect syntactic dependencies. Our results on benchmark datasets demonstrate that our model outperforms the previous methods and achieves state-of-the-art results for the ALSC task.

Multitask learning has shown promising performance in learning multiple related tasks simultaneously, and variants of model architectures have been proposed, especially for supervised classification problems. One goal of multitask learning is to extract a good representation that sufficiently captures the relevant part of the input about the output for each learning task. To achieve this objective, in this paper we design a multitask learning architecture based on the observation that correlations exist between outputs of some related tasks (e.g. entity recognition and relation extraction tasks), and they reflect the relevant features that need to be extracted from the input. As outputs are unobserved, our proposed model exploits task predictions in lower layers of the neural model, also referred to as early predictions in this work. But we control the injection of early predictions to ensure that we extract good task-specific representations for classification. We refer to this model as a Progressive Multitask learning model with Explicit Interactions (PMEI). Extensive experiments on multiple benchmark datasets produce state-of-the-art results on the joint entity and relation extraction task.

Attention with positional encoding has been demonstrated as a powerful component in modern neural network models, such as transformers. However, why positional encoding works well in attention models remains largely unanswered. In this paper, we study the scalar relative positional encoding (SRPE) proposed in the T5 transformer. Such an encoding method has two features. First, it uses a scalar to embed relative positions. Second, the relative positions are bucketized using a fixed heuristic algorithm, and positions in the same bucket share the same embedding. In this work, we show that SRPE in attention has an elegant probabilistic interpretation. More specifically, the positional encoding serves to produce a prior distribution for the attended positions. The resulting attentive distribution can be viewed as a posterior distribution of the attended position given the observed input sequence. Furthermore, we propose a new SRPE (AT5) that adopts a learnable bucketization protocol and automatically adapts to the dependency range specific to the learning task. Empirical studies show that the AT5 achieves superior performance than the T5's SRPE.

Effective regularization techniques are highly desired in deep learning for alleviating overfitting and improving generalization. This work proposes a new regularization scheme, based on the understanding that the flat local minima of the empirical risk cause the model to generalize better. This scheme is referred to as adversarial model perturbation (AMP), where instead of directly minimizing the empirical risk, an alternative "AMP loss" is minimized via SGD. Specifically, the AMP loss is obtained from the empirical risk by applying the "worst" norm-bounded perturbation on each point in the parameter space. Comparing with most existing regularization schemes, AMP has strong theoretical justifications, in that minimizing the AMP loss can be shown theoretically to favour flat local minima of the empirical risk. Extensive experiments on various modern deep architectures establish AMP as a new state of the art among regularization schemes. Our code is available at https://github.com/hiyouga/AMP-Regularizer.

Fine-grained entity typing (FET) aims to annotate the entity mentions in a sentence with fine-grained type labels. It brings plentiful semantic information for many natural language processing tasks. Existing FET approaches apply hard attention to learn on the noisy labels, and ignore that those noises have structured hierarchical dependency. Despite their successes, these FET models are insufficient in modeling type hierarchy dependencies and handling label noises. In this paper, we directly tackle the structured noisy labels by combining a forward tree module and a backward tree module. Specifically, the forward tree formulates the informative walk that hierarchically represents the type distributions. The backward tree models the erroneous walk that learns the noise confusion matrix. Empirical studies on several benchmark data sets confirm the effectiveness of the proposed framework.

In this paper, we unify a diverse set of learning tasks in NLP, semantic retrieval and related areas, under a common umbrella, which we call unsupervised semantic association learning (USAL). Examples of this generic task include word sense disambiguation, answer selection and question retrieval. We then present a novel modeling framework to tackle such tasks. The framework introduces, under the deep learning paradigm, a latent label indexing the true target in the candidate target set. An EM algorithm is then developed for learning the deep model and inferring the latent variables, principled under variational techniques and noise contrastive estimation. We apply the model and algorithm to several semantic retrieval benchmark tasks and the superior performance of the proposed approach is demonstrated via empirical studies.

We consider the problem of learning a neural network classifier. Under the information bottleneck (IB) principle, we associate with this classification problem a representation learning problem, which we call “IB learning”. We show that IB learning is, in fact, equivalent to a special class of the quantization problem. The classical results in rate-distortion theory then suggest that IB learning can benefit from a “vector quantization” approach, namely, simultaneously learning the representations of multiple input objects. Such an approach assisted with some variational techniques, result in a novel learning framework, “Aggregated Learning”, for classification with neural network models. In this framework, several objects are jointly classified by a single neural network. The effectiveness of this framework is verified through extensive experiments on standard image recognition and text classification tasks.

A large majority of approaches have been proposed to leverage the dependency tree in the relation classification task. Recent works have focused on pruning irrelevant information from the dependency tree. The state-of-the-art Attention Guided Graph Convolutional Networks (AGGCNs) transforms the dependency tree into a weighted-graph to distinguish the relevance of nodes and edges for relation classification. However, in their approach, the graph is fully connected, which destroys the structure information of the original dependency tree. How to effectively make use of relevant information while ignoring irrelevant information from the dependency trees remains a challenge in the relation classification task. In this work, we learn to transform the dependency tree into a weighted graph by considering the syntax dependencies of the connected nodes and persisting the structure of the original dependency tree. We refer to this graph as a syntax-transport graph. We further propose a learnable syntax-transport attention graph convolutional network (LST-AGCN) which operates on the syntax-transport graph directly to distill the final representation which is sufficient for classification. Experiments on Semeval-2010 Task 8 and Tacred show our approach outperforms previous methods.

Aspect-level sentiment classification (ALSC) aims at predicting the sentiment polarity of a specific aspect term occurring in a sentence. This task requires learning a representation by aggregating the relevant contextual features concerning the aspect term. Existing methods cannot sufficiently leverage the syntactic structure of the sentence, and hence are difficult to distinguish different sentiments for multiple aspects in a sentence. We perceive the limitations of the previous methods and propose a hypothesis about finding crucial contextual information with the help of syntactic structure. For this purpose, we present a neural network model named RepWalk which performs a replicated random walk on a syntax graph, to effectively focus on the informative contextual words. Empirical studies show that our model outperforms recent models on most of the benchmark datasets for the ALSC task. The results suggest that our method for incorporating syntactic structure enriches the representation for the classification.

The dependencies between system and user utterances in the same turn and across different turns are not fully considered in existing multidomain dialogue state tracking (MDST) models.In this study, we argue that the incorporation of these dependencies is crucial for the design of MDST and propose Parallel Interactive Networks (PIN) to model these dependencies.Specifically, we integrate an interactive encoder to jointly model the in-turn dependencies and cross-turn dependencies.The slot-level context is introduced to extract more expressive features for different slots.And a distributed copy mechanism is utilized to selectively copy words from historical system utterances or historical user utterances.Empirical studies demonstrated the superiority of the proposed PIN model.

The idea of using multi-task learning approaches to address the joint extraction of entity and relation is motivated by the relatedness between the entity recognition task and the relation classification task.Existing methods using multi-task learning techniques to address the problem learn interactions among the two tasks through a shared network, where the shared information is passed into the taskspecific networks for prediction.However, such an approach hinders the model from learning explicit interactions between the two tasks to improve the performance on the individual tasks.As a solution, we design a multitask learning model which we refer to as recurrent interaction network which allows the learning of interactions dynamically, to effectively model task-specific features for classification.Empirical studies on two real-world datasets confirm the superiority of the proposed model.

The introduction of VAE provides an efficient framework for the learning of generative models, including generative topic models.However, when the topic model is a Latent Dirichlet Allocation (LDA) model, a central technique of VAE, the reparameterization trick, fails to be applicable.This is because no reparameterization form of Dirichlet distributions is known to date that allows the use of the reparameterization trick.In this work, we propose a new method, which we call Rounded Reparameterization Trick (RRT), to reparameterize Dirichlet distributions for the learning of VAE-LDA models.This method, when applied to a VAE-LDA model, is shown experimentally to outperform the existing neural topic models on several benchmark datasets and on a synthetic dataset.

Entity linking, which maps named entity mentions in a document into the proper entities in a given knowledge graph, has been shown to be able to significantly benefit from modeling the entity relatedness through Graph Convolutional Networks (GCN). Nevertheless, existing GCN entity linking models fail to take into account the fact that the structured graph for a set of entities not only depends on the contextual information of the given document but also adaptively changes on different aggregation layers of the GCN, resulting in insufficiency in terms of capturing the structural information among entities. In this paper, we propose a dynamic GCN architecture to effectively cope with this challenge. The graph structure in our model is dynamically computed and modified during training. Through aggregating knowledge from dynamically linked nodes, our GCN model can collectively identify the entity mappings between the document and the knowledge graph, and efficiently capture the topical coherence among various entity mentions in the entire document. Empirical studies on benchmark entity linking data sets confirm the superior performance of our proposed strategy and the benefits of the dynamic graph structure.

Supervised learning relies heavily on readily available labelled data to infer an effective classification function. However, proposed methods under the supervised learning paradigm are faced with the scarcity of labelled data within domains, and are not generalized enough to adapt to other tasks. Transfer learning has proved to be a worthy choice to address these issues, by allowing knowledge to be shared across domains and tasks. In this paper, we propose two transfer learning methods Anchored Model Transfer (AMT) and Soft Instance Transfer (SIT), which are both based on multi-task learning, and account for model transfer and instance transfer, and can be combined into a common framework. We demonstrate the effectiveness of AMT and SIT for aspect-level sentiment classification showing the competitive performance against baseline models on benchmark datasets. Interestingly, we show that the integration of both methods AMT+SIT achieves state-of-the-art performance on the same task.

Real-world knowledge bases such as DBPedia, Yago, and Freebase contain sparse linkage connectivity, which poses a severe challenge to link prediction between entities. To cope with such data scarcity issues, recent models have focused on learning interactions between entity pairs by means of relations that exist between them. However promising, some relations are associated with very few tail entities or head entities, resulting in poor estimation of the relation interaction between entities. In this article, we break the sole dependency of modeling relation interactions between entity pairs by associating a triple with pairwise embeddings, i.e., distributed vector representations for pairs of word-based entities and relation of a triple. We capture the interactions that exist between pairwise embeddings by means of a Pairwise Factorization Model that employs a factorization machine with relation attention. This approach allows parameters for related interactions to be estimated efficiently, ensuring that the pairwise embeddings are discriminative, providing strong supervisory signals for the decoding task of link prediction. The Pairwise Factorization Model we propose exploits a neural bag-of-words model as the encoder, which effectively encodes word-based entities into distributed vector representations for the decoder. The proposed model is simple and enjoys efficiency and capability, showing superior link prediction performance over state-of-the-art complex models on benchmark datasets DBPedia50K and FB15K-237.

MixUp (Zhang et al. 2017) is a recently proposed dataaugmentation scheme, which linearly interpolates a random pair of training examples and correspondingly the one-hot representations of their labels. Training deep neural networks with such additional data is shown capable of significantly improving the predictive accuracy of the current art. The power of MixUp, however, is primarily established empirically and its working and effectiveness have not been explained in any depth. In this paper, we develop an understanding for MixUp as a form of “out-of-manifold regularization”, which imposes certain “local linearity” constraints on the model’s input space beyond the data manifold. This analysis enables us to identify a limitation of MixUp, which we call “manifold intrusion”. In a nutshell, manifold intrusion in MixUp is a form of under-fitting resulting from conflicts between the synthetic labels of the mixed-up examples and the labels of original training data. Such a phenomenon usually happens when the parameters controlling the generation of mixing policies are not sufficiently fine-tuned on the training data. To address this issue, we propose a novel adaptive version of MixUp, where the mixing policies are automatically learned from the data using an additional network and objective function designed to avoid manifold intrusion. The proposed regularizer, AdaMixUp, is empirically evaluated on several benchmark datasets. Extensive experiments demonstrate that AdaMixUp improves upon MixUp when applied to the current art of deep classification models.

Embedding based models for knowledge base completion have demonstrated great successes and attracted significant research interest. In this work, we observe that existing embedding models all have their loss functions decomposed into atomic loss functions, each on a triple or an postulated edge in the knowledge graph. Such an approach essentially implies that conditioned on the embeddings of the triple, whether the triple is factual is independent of the structure of the knowledge graph. Although arguably the embeddings of the entities and relation in the triple contain certain structural information of the knowledge base, we believe that the global information contained in the embeddings of the triple can be insufficient and such an assumption is overly optimistic in heterogeneous knowledge bases. Motivated by this understanding, in this work we propose a new embedding model in which we discard the assumption that the embeddings of the entities and relation in a triple is a sufficient statistic for the triple’s factual existence. More specifically, the proposed model assumes that whether a triple is factual depends not only on the embedding of the triple but also on the embeddings of the entities and relations in a larger graph neighbourhood. In this model, attention mechanisms are constructed to select the relevant information in the graph neighbourhood so that irrelevant signals in the neighbourhood are suppressed. Termed locality-expanded neural embedding with attention (LENA), this model is tested on four standard datasets and compared with several stateof-the-art models for knowledge base completion. Extensive experiments suggest that LENA outperforms the existing models in virtually every metric.

Ci is a lyric poetry form that follows highly restrictive metrical structures. This makes it challenging for a computer to compose Ci subject to a specified metrical requirement. In this work, we adapt the CVAE framework to automated Ci generation under metrical constraints. Specifically, we present the first neural model that explicitly encodes the designated metrical structure for Ci generation. The proposed model is shown experimentally to generate Ci with nearly perfect metrical structures.

Junfan Chen, Richong Zhang, Yongyi Mao, Hongyu Guo, Jie Xu. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.

Kai Sun, Richong Zhang, Samuel Mensah, Yongyi Mao, Xudong Liu. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.

Fine-grained entity typing (FET), which annotates the entities in a sentence with a set of finely specified type labels, often serves as the first and critical step towards many natural language processing tasks. Despite great processes have been made, current FET methods have difficulty to cope with the noisy labels which naturally come with the data acquisition processes. Existing FET approaches either pre-process to clean the noise or simply focus on one of the noisy labels, sidestepping the fact that those noises are related and content dependent. In this paper, we directly model the structured, noisy labels with a novel content-sensitive weighting schema. Coupled with a newly devised cost function and a hierarchical type embedding strategy, our method leverages a random walk process to effectively weight out noisy labels during training. Experiments on several benchmark datasets validate the effectiveness of the proposed framework and establish it as a new state of the art strategy for noisy entity typing problem.

Knowledge graphs such as DBPedia and Freebase contain sparse linkage connectivity, which poses severe challenge to link prediction between entities. In addressing this sparsity problem, our studies indicate that one needs to leverage model with low complexity to avoid overfitting the weak structural information in the graphs, requiring the simple models which can efficiently encode the entities and their description information and then effectively decode their relationships. In this paper, we present a simple and efficient model that can attain these two goals. Specifically, we use a bag-of-words model, where relevant words are aggregated using average pooling or a basic Graph Convolutional Network to encode entities into distributed embeddings. A factorization machine is then used to score the relationships between those embeddings to generate linkage predictions. Empirical studies on two real datasets confirms the efficiency of our proposed model and shows superior predictive performance over state-of-the-art approaches.

Microblog platforms have been extremely popular in the big data era due to its real-time diffusion of information. It's important to know what anomalous events are trending on the social network and be able to monitor their evolution and find related anomalies. In this paper we demonstrate RING, a real-time emerging anomaly monitoring system over microblog text streams. RING integrates our efforts on both emerging anomaly monitoring research and system research. From the anomaly monitoring perspective, RING proposes a graph analytic approach such that (1) RING is able to detect emerging anomalies at an earlier stage compared to the existing methods, (2) RING is among the first to discover emerging anomalies correlations in a streaming fashion, (3) RING is able to monitor anomaly evolutions in real-time at different time scales from minutes to months. From the system research perspective, RING (1) optimizes time-ranged keyword query performance of a full-text search engine to improve the efficiency of monitoring anomaly evolution, (2) improves the dynamic graph processing performance of Spark and implements our graph stream model on it, As a result, RING is able to process big data to the entire Weibo or Twitter text stream with linear horizontal scalability. The system clearly presents its advantages over existing systems and methods from both the event monitoring perspective and the system perspective for the emerging event monitoring task.

Question answering over knowledge bases aims to take full advantage of the information in knowledge bases with the ultimate purpose of returning answers to questions. To access the substantial knowledge within the KB, many model architectures are hindered by the bottleneck of accurately predicting relations that connect subject entities in questions to object entities in the knowledge base. To break the bottleneck, this article presents a novel model architecture, APVA, which includes a verification mechanism to check the correctness of predicted relations. Specifically, APVA takes advantage of KB-based information to improve relation prediction but verifies the correctness of the predicted relation by means of simple negative sampling in a logistic regression framework. The APVA architecture offers a natural way to integrate an iterative training procedure, which we call turbo training. Accordingly, we introduce APVA-TURBO to perform question answering over knowledge bases. We demonstrate extensive experiments to show that APVA-TURBO outperforms existing approaches on question answering.

Embedding has emerged as an important approach to prediction, inference, data mining and information retrieval based on knowledge bases and various embedding models have been presented. Most of these models are "typeless," namely, treating a knowledge base solely as a collection of instances without considering the types of the entities therein. In this paper, we investigate the use of entity type information for knowledge base embedding. We present a framework that augments a generic "typeless" embedding model to a typed one. The framework interprets an entity type as a constraint on the set of all entities and let these type constraints induce isomorphically a set of subsets in the embedding space. Additional cost functions are then introduced to model the fitness between these constraints and the embedding of entities and relations. A concrete example scheme of the framework is proposed. We demonstrate experimentally that this framework offers improved embedding performance over the typeless models and other typed models.

In this paper, we study the problem of question answering over knowledge base. We identify that the primary bottleneck in this problem is the difficulty in accurately predicting the relations connecting the subject entity to the object entities. We advocate a new model architecture, APVA, which includes a verification mechanism responsible for checking the correctness of predicted relations. The APVA framework naturally supports a well-principled iterative training procedure, which we call turbo training. We demonstrate via experiments that the APVA-TUBRO approach drastically improves the question answering performance.

We propose a novel strategy to encode the syntax parse tree of sentence into a learnable distributed representation.The proposed syntax encoding scheme is provably informationlossless.In specific, an embedding vector is constructed for each word in the sentence, encoding the path in the syntax tree corresponding to the word.The one-to-one correspondence between these "syntax-embedding" vectors and the words (hence their embedding vectors) in the sentence makes it easy to integrate such a representation with all word-level NLP models.We empirically show the benefits of the syntax embeddings on the Authorship Attribution domain, where our approach improves upon the prior art and achieves new performance records on five benchmarking data sets.

Building knowledge base embedding models for link prediction has achieved great success. We however argue that the conventional top-k criterion used for evaluating the model performance is inappropriate. This paper introduces a new criterion, referred to as max-k. Through theoretical analysis and experimental study, we show that the top-k criterion is fundamentally inferior to max-k. We also introduce two prediction protocols for the max-k criterion. These protocols are strongly justified theoretically. Various insights concerning the max-k criterion and the two protocols are obtained through extensive experiments.

The knowledge base (KB) completion problem is usually formulated as a link prediction problem. Such formulation is incapable of capturing certain application scenarios when the KB contains multi-fold relations. In this paper, we present a new formulation of KB completion, called instance reconstruction. Unlike its link-prediction counterpart, which has linear complexity in the size of the KB, this problem has its complexity behave as a high-degree polynomial. This presents a significant challenge in developing scalable instance reconstruction algorithms. In this paper, we present a novel knowledge embedding model (RAE) and build on it an instance reconstruction algorithm (SIR). The SIR algorithm utilizes schema-based filtering as well as "relatedness" filtering for complexity reduction. Here relatedness refers to the likelihood that two entities co-participate in a common instance, and the relatedness metric is learned from the RAE model. We show experimentally that SIR significantly reduces computation complexity without sacrificing reconstruction performance. The complexity reduction corresponds to reducing the KB size by 100 to 1000 folds.

User interest prediction plays an important role in online services, such as electronic commerce, social network, and online advertising. This information is not always explicitly available for online systems. To identify the interests, existing studies merely focused on modeling the relationships between user generated contents and user interests. However, the interests of a user should not only be inferred by user generated contents, but also the relationships between users which might imply the user interests. In this paper, our goal is to unveil the true interests of users based on user generated contents as well as relationships between users. We built a probabilistic user interests model and proposed a user interests propagation algorithm (UIP) to tackle this problem. A factor graph-based approach is utilized to estimate the distribution of the interests of users. We conducted experiments on real-world datasets to validate the effectiveness of the model. Furthermore, we integrated our UIP algorithm with the classical matrix factorization algorithm to deal with the rating prediction task. Experimental studies confirm the superiority of the proposed approach.

This paper studies the problem of designing practical web service recommender systems that make quality predictions based on web service invocation histories. We formalize a generic architecture for such systems, that include both an online training module and an off-line training module. In addition, we develop a general online and off-line training algorithms to demonstrate the advantages of such an architecture and the natural fit of stochastic gradient descent algorithms for the architecture. The advantages of our proposed architecture has been confirmed by the comparisons with existing web service recommendation algorithms on a real-life dataset.

Summary As a new form of social media, microblogs (e.g., Twitter and Weibo) are playing an important role in people's daily life. With the rise in popularity and size of microblogs, there is a need for distributed approaches that can detect bursty event with low latency from the short‐text data stream. In this paper, we propose a distributed and incremental temporal topic model for microblogs called Bursty Event dEtection (BEE+). BEE+ is able to detect bursty events from short‐text dataset and model the temporal information. And BEE+ processes the post‐stream incrementally to track the topic drifting of events over time. Therefore, the latent semantic indices are preserved from one time period to the next. In order to achieve real‐time processing, we design a distributed execution framework based on Spark engine. To verify its ability to detect bursty event, we conduct experiments on a Weibo dataset of 6,360,125 posts. The results show that BEE+ can outperform the baselines for detecting the meaningful bursty events and track the topic drifting. Copyright © 2015 John Wiley & Sons, Ltd.

One of the most important crowdsourcing topics is to study the effective quality control methods so as to reduce the cost and to guarantee the quality of task processing. As an effective approach, iterative improvement workflow is known to choose the best result from multiple workflows. However, for complex crowdsourcing tasks that consists of a certain number of subtasks under some specific constraints, but cannot be split into subtasks to be crowdsourced, the approach merely considers the best workflow without integrating the contributions of all workflows, which potentially results in extra costs for more iterations. In this paper, we propose an assembly model to integrate the best output of subtasks from different workflows. Moreover, we devise an efficient iterative method based on POMDP to improve the quality of assembled output. Empirical studies confirms the superiority of our proposed model.

Mashing up Web services and RESTful APIs is a novel programming approach to develop new applications. As the number of available resources is increasing rapidly, to discover potential services or APIs is getting difficult. Therefore, it is vital to relieve mashup developers of the burden of service discovery. In this paper, we propose a probabilistic model to assist mashup creators by recommending a list of APIs that may be used to compose a required mashup given descriptions of the mashup. Specifically, a relational topic model is exploited to characterize the relationship among mashups, APIs and their links. In addition, we incorporate the popularity of APIs to the model and make predictions on the links between mashups and APIs. Moreover, the statistical analysis on a public mashup platform shows the current status of mashup development and the applicability of this study. Experiments on a large service data set confirm the effectiveness of this proposed approach.

The prediction of trust relationships in social networks plays an important role in the analytics of the networks. Although various link prediction algorithms for general networks may be adapted for this purpose, the recent notion of “trust propagation” has been shown to effectively capture the trust-formation mechanisms and resulted in an effective prediction algorithm. This article builds on the concept of trust propagation and presents a probabilistic trust propagation model. Our model exploits the modern framework of probabilistic graphical models, more specifically, factor graphs. Under this model, the trust prediction problem can be formulated as a statistical inference problem and we derive the belief propagation algorithm as a solver for trust prediction. The model and algorithm are tested using datasets from Epinions and Ciao, by which performance advantages over the previous algorithms are demonstrated.

MapReduce is a widely used programming model for large scale data processing. In order to estimate the performance of MapReduce job and analyze the bottleneck of MapReduce job, a practical performance model for MapReduce is needed. Many works have been done on modeling the performance of MapReduce jobs. However, existing performance models ignore some important factors, such as I/O congestions and task failures over cluster, which may significantly change the execution costs of MapReduce job. This paper, aiming at predicting the execution time of a MapReduce job, presents an enhanced performance model that takes the resource contention and task failures into consideration. In addition, the experimental results show that the model is more accurate than those without considering the contention and failure factors.

As Internet is changing from a network of data into a network of functionalities, a federated Internet of applications is a natural trending topic, where every application can cooperate with each other smoothly to serve users. Cooperation can be regarded as integrating applications of different providers for users. However, existing integration techniques do not pay enough attention to multiple participants including application providers and end-users. In this study, we advocate a global cooperation model of Internet of applications for all the participants. Specifically, we propose an intermediary based model to realize the cooperation among applications. With this model, on the one hand, users can be greatly facilitated to cooperatively use applications from various providers to meet their individualized requirements; on the other hand, providers can easily enable their own applications to interact with those from other providers. Thus, the federated Internet of applications is easier to be achieved than using existing solutions. In addition, we implement the model and show some case studies which demonstrate the effectiveness of this model. In our vision, such a model is the beginning to bring the socialized relationships behind applications into the digital world.

The increasing availability of socially shared media with tags annotated makes it vital for retrieval approaches to precisely detect web content topic semantic and better understand user interest. Most existing methodologies process the queries merely considering user posted keywords and retrieve media labeled with tags that are similar to query words, while ignoring users implicit interests and preferences. This fact stimulates us to develop preference discovering models to reveal the users' latent intents. In this paper, we study the problem of finding user preference and interest from folksonomy corpus and propose a preference-topic model that exploits probabilistic graphical model and Gibbs sampling algorithm to infer the user interested latent semantic topics. The experimental results show that, with the help of the proposed model, preference topics of the web content creators can be effectively discovered. In addition, two exemplified applications are discussed briefly.

This paper presents an global social travel system to assist tourists in their itinerary planning, tour navigation, and travel knowledge sharing. In particular, firstly, we propose an efficient and flexible itinerary planning algorithm for organizing itineraries. Secondly, we design an intelligent tour path planning and navigation system by mining patterns from trajectories and geo-photos shared by other tourists. Thirdly, we provide a framework to monitor the status and events that may affect the predefined itineraries. Finally, the social media module enhances the opportunities of experience (itinerary, trajectory, and travelogue) sharing and helps travelers make better decisions. Concrete demonstrations of our system are also provided in this paper to show the flexibility and applicability of our system.

SUMMARY This paper studies the problem of designing real‐time helpfulness prediction algorithms. Instead of following the conventional route, in which the fraction of positive votes is used as the measure of helpfulness, we give ‘helpfulness’ a naturally sensible and mathematically precise definition, namely, as the probability that a user will vote ‘helpful’ on the user‐generated content. Building on this definition, we introduce a principled methodology to helpfulness prediction, in which the prediction problem is naturally formulated as an optimization problem. Under this proposed methodology, we first develop a batch (off‐line) algorithm. Experiments on data from Amazon.com suggest that our proposed model in fact outperforms the previously reported prediction algorithm, support vector regression. In some circumstances, an online algorithm that can update the model as additional data arrive is required. In light of this, we proposed an online algorithm that incrementally updates the parameters of the model. Finally, an efficient hybrid algorithm is provided to increase the convergence rate and prediction precision. The final two algorithms are tested on real‐life user‐generated contents, and experimental results illustrate that the hybrid approach efficiently processes incoming data and generates reliable helpfulness predictions for users. Copyright © 2011 John Wiley & Sons, Ltd.

Electronic Word-of-Mouth (eWOM) is growing exponentially with the rapid development of electronic commerce. As a result, consumers are increasingly crowded by a huge amount of eWOM contents and therefore there is a need to automatically recommend eWOM contents that are helpful to them. Existing helpfulness assessment approaches that deterministically estimate the helpfulness of eWOM contents lack a generative formulation and are limited to the training set that has been voted by many readers. This article presents a rigorous probabilistic framework for inferring the “helpfulness” of eWOM contents which can build a “helpfulness” model from a low number of votes on eWOM contents. Furthermore, we introduce a measurement, “helpfulness” bias, as the benchmark for the “helpfulness” of eWOM documents. We also propose a model that exploits the graphical model and expectation maximization algorithm, under this probabilistic framework, to demonstrate the versatility of our framework. Our algorithm is compared experimentally to other existing helpfulness discovering algorithms and the experimental results show that our framework can effectively model the helpfulness of eWOM contents better than other approaches, and therefore indicate the capability of our framework to recommend helpful eWOMs to potential consumers.

User-generated reviews play an important role for potential consumers in making purchase decisions. However, the quality and helpfulness of user-generated reviews are unavailable unless consumers read through them. Automatically predicting the helpfulness of user-generated reviews can assist consumers in discovering helpful reviews. Existing helpfulness assessing models make use of the positive vote fraction as a benchmark and focus on heuristically finding a ``best guess'' value, which is a point estimate of helpfulness. This benchmark methodology ignores the voter population size and the uncertainty of the helpfulness estimation. In this paper, we propose a user-generated review recommendation model based on the probability density of the review's helpfulness, rather than using the positive vote fraction. Our proposed model exploits probabilistic methodology to infer the helpfulness distribution. Furthermore, our experimental results confirm that our approach can effectively assess the helpfulness of user-generated reviews and recommend the most helpful ones to consumers.

Automatically assessing the quality and helpfulness of consumer reviews is more and more desirable with the evolutionary development of online review systems. Existing helpfulness assessment methodologies make use of the positive vote fraction as a benchmark and heuristically find a "best guess" to estimate the helpfulness of review documents. This benchmarking methodology ignores the voter population size and treats the the same positive vote fraction as the same helpfulness value. We propose a review recommendation approach that make use of the probability density of the review helpfulness as the benchmark and exploit graphical model and Expectation Maximization (EM) algorithm for the inference of review helpfulness. The experimental results demonstrate that the proposed approach is superior to existing approaches.

E-commerce Web sites, such as Amazon.com, provide platforms for consumers to review products and share their opinions. However, it is impossible for consumers to read throughout the huge amount of available reviews. In addition, the quality and helpfulness of reviews are unavailable unless consumers have to read through them.This paper proposes an entropy-based model to predict the helpfulness of reviews. Reviews can be ranked by our entropy-based scoring model and reviews that may help consumers better than others will be found. We also compare our model with several machine learning algorithms. Our experimental results show that our approach is effective in ranking and classifying online reviews. With the predicted helpfulness of reviews, consumers can make purchase decisions more easily.