Repository-level code generation has attracted growing interest, yet most benchmarks and methods remain maintainer-centric, emphasizing bug fixing and feature implementation. In contrast, a common yet underexplored scenario is repository usage: external users want to build applications by correctly invoking repository-internal APIs, composing them into runnable end-to-end workflows rather than modifying the codebase. To support this setting, we introduce RUCCE, a benchmark for repository-usage code generation built from real-world Python repositories. Each instance pairs a natural-language usage instruction with grounded target APIs and a verified reference script, enabling evaluation of both API retrieval and repository-usage code generation. Building on RUCCE, we propose RUCACoder, a closed-loop multi-agent framework with a Retriever for hierarchical repository exploration, a Verifier for reranking and validation, and a Coder for feedback-driven script synthesis. Experiments across multiple backbone LLMs show that RUCACoder consistently outperforms strong retrieval and generation baselines.

In this paper, we address a novel problem in social network analysis: the Most Influenced Community Search (MICS). Given a graph and a seed node set$S$, the MICS problem seeks to identify a densely connected sub graph that is most significantly impacted by$S$. We formally define MICS, prove its NP-hardness, and show that constant-factor approximation is not feasible. To solve MICS efficiently, we propose a two-phase framework. In the first phase, we compute the influenced expectation for each node, representing its likelihood of being influenced by$S$. We develop two algorithms: S-InfExp, a sampling-based method with theoretical guarantees, and L-InfExp, a learning-based approach for faster predictions. In the second phase, we introduce two algorithms, GlobalSearch and LocalSearch, to find the most influenced community. GlobalSearch uses a top-down, greedy approach, while LocalSearch applies a bottom-up strategy. Experiments on eight real-world datasets demonstrate that (1) L-InfExp is up to 100× faster than S-InfExp with comparable accuracy, (2) LocalSearch is 10× faster than GlobalSearch, with both algorithms effectively identifying the community with the highest influenced expectations, and (3) our algorithms outperform all baselines.

Joint multimodal entity-relation extraction (JMERE) is a challenging task that involves two joint subtasks, i.e., named entity recognition and relation extraction, from multimodal data such as text sentences with associated images. Previous JMERE methods have primarily employed 1) pipeline models, which apply pre-trained unimodal models separately and ignore the interaction between tasks, or 2) word-pair relation tagging methods, which neglect neighboring word pairs. To address these limitations, we propose a fine-grained network for JMERE. Specifically, we introduce a fine-grained alignment module that utilizes a phrase-patch to establish connections between text phrases and visual objects. This module can learn consistent multimodal representations from multimodal data. Furthermore, we address the task-irrelevant image information issue by proposing a gate fusion module, which mitigates the impact of image noise and ensures a balanced representation between image objects and text representations. Furthermore, we design a multi-word decoder that enables ensemble prediction of tags for each word pair. This approach leverages the predicted results of neighboring word pairs, improving the ability to extract multi-word entities. Evaluation results from a series of experiments demonstrate the superiority of our proposed model over state-of-the-art models in JMERE.

In the evolving field of Natural Language Processing (NLP), understanding the temporal context of text is increasingly critical for applications requiring advanced temporal reasoning. Traditional pre-trained language models like BERT, which rely on synchronic document collections such as BookCorpus and Wikipedia, often fall short in effectively capturing and leveraging temporal information. To address this limitation, we introduce BiTimeBERT 2.0, a novel time-aware language model pre-trained on a temporal news article collection. BiTimeBERT 2.0 incorporates temporal information through three innovative pre-training objectives: Extended Time-Aware Masked Language Modeling (ETAMLM), Document Dating (DD), and Time-Sensitive Entity Replacement (TSER). Each objective is specifically designed to target a distinct dimension of temporal information: ETAMLM enhances the model’s understanding of temporal contexts and relations, DD integrates document timestamps as explicit chronological markers, and TSER focuses on the temporal dynamics of “Person” entities. Moreover, our refined corpus preprocessing strategy reduces training time by nearly 53%, making BiTimeBERT 2.0 significantly more efficient while maintaining high performance. Experimental results show that BiTimeBERT 2.0 achieves substantial improvements across a broad range of time-related tasks and excels on datasets spanning extensive temporal ranges. These findings underscore BiTimeBERT 2.0’s potential as a powerful tool for advancing temporal reasoning in NLP. 1

Most existing methods for visual domain adaptation need to convert high-order tensors into one-order high-dimensional vectors through naive vectorization operations. However, they not only destroy the internal spatial structure within the original high-order tensors, but also result in exponentially increasing model parameters. To address these problems, we propose a novel method for visual domain adaptation by representing tensorial features in tensor-train subspace in this paper. Specifically, we firstly provide a theoretical deduction by constructing a tensor-train subspace and proving its linearity and left-orthogonality. Secondly, to extract common tensorial features between source and target domains, we formulate the visual domain adaptation problem into an optimization problem that models the aforementioned common tensor-train subspace between two domains, as well as their corresponding projections. Thirdly, we design a tensor-train subspace representation algorithm (TTSR) to solve the multiple variables optimization problem by optimizing its sub-problems iteratively, so as to process high-order tensorial features. Finally, we evaluate the performance of our proposed TTSR algorithm by conducting extensive experiments on three popular public datasets. The experimental results demonstrate that the TTSR algorithm can improve the classification accuracy of unlabeled target domain than that of baseline algorithms.

Benefiting many information retrieval applications, named entity recognition (NER) has shown impressive progress. Recently, there has been a growing trend to decompose complex NER tasks into two subtasks (e.g., entity span detection (ESD) and entity type classification (ETC), to achieve better performance. Despite the remarkable success, from the perspective of representation, existing methods do not explicitly distinguish non-entities and entities, which may lead to ESD errors. Meanwhile, they do not explicitly distinguish entities with different entity types, which may lead to entity type misclassification. As such, the limited representation abilities may challenge some competitive NER methods, leading to unsatisfactory performance, especially in the low-resource setting (e.g., cross-domain NER). In light of these challenges, we propose to utilize contrastive learning to refine the original chaotic representations and learn the generalized representations for cross-domain NER. In particular, this article proposes a dual contrastive learning model (Dual-CL), which respectively utilizes a token-level contrastive learning module and a sentence-level contrastive learning module to enhance ESD, ETC for cross-domain NER. Empirical results on 10 domain pairs under two different settings show that Dual-CL achieves better performances than compared baselines in terms of several standard metrics. Moreover, we conduct detailed analyses to are presented to better understand each component’s effectiveness.

Time is an important aspect of documents and is used in a range of NLP and IR tasks. In this work, we investigate methods for incorporating temporal information during pre-training to further improve the performance on time-related tasks. Compared with common pre-trained language models like BERT which utilize synchronic document collections (e.g., BookCorpus and Wikipedia) as the training corpora, we use long-span temporal news article collection for building word representations. We introduce BiTimeBERT, a novel language representation model trained on a temporal collection of news articles via two new pre-training tasks, which harnesses two distinct temporal signals to construct time-aware language representations. The experimental results show that BiTimeBERT consistently outperforms BERT and other existing pre-trained models with substantial gains on different downstream NLP tasks and applications for which time is of importance (e.g., the accuracy improvement over BERT is 155% on the event time estimation task).

To address the scarcity of massive labeled data, cross-domain named entity recognition (cross-domain NER) attracts increasing attention. Recent studies focus on decomposing NER into two separate tasks (i.e., entity span detection and entity type classification) to reduce the complexity of the cross-domain transfer. Despite the promising results, there still exists room for improvement. In particular, the rich domain-shared syntactic and semantic information, which are respectively important for entity span detection and entity type classification, are still underutilized. In light of these two challenges, we propose applying graph attention networks (GATs) to encode the above two kinds of information. Moreover, considering that GATs mainly operate in the Euclidean space, which may fail to capture the latent hierarchical relations among words for learning high-quality word representations, we further propose to embed words into Hyperbolic spaces. Finally, a decouple hyperbolic graph attention network (DH-GAT) is introduced for cross-domain NER. Empirical results on 10 domain pairs show that DH-GAT achieves state-of-the-art performance on several standard metrics, and further analyses are presented to better understand each component's effectiveness.

Relation Classification is one of the most important tasks in text mining. Previous methods either require large-scale manually-annotated data or rely on distant supervision approaches which suffer from the long-tail problem. To reduce the expensive manually-annotating cost and solve the long-tail problem, prototypical networks are widely used in few-shot RC tasks. Despite their remarkable performance, current prototypical networks ignore the different granularities of relations, which degrades the classification performance dramatically. Moreover, the optimization of current prototypical networks simply relies on the cross-entropy loss, which cannot consider the intra-relation compactness and the dispersion among relations in a semantic space. It is not robust enough for current prototypical network in real-world and complicated scenarios. In this paper, we propose an area prototypical network with a granularity-aware measurement, aiming to considering the different granularities of relations. Each relation is represented as an area whose width can reflect the granularity level of relation. Moreover, to improve the robustness, bimargin loss is designed to force area prototypical network to improve the intra-relation compactness and inter-relation dispersion for the feature representation in a semantic space. Extensive experiments on two public datasets are conducted and evaluate the effectiveness of our proposed model.

The notion of word embedding plays a fundamental role in natural language processing (NLP). However, pre-training word embedding for very large-scale vocabulary is computationally challenging for most existing methods. In this work, we show that with merely a small fraction of contexts (Q-contexts) which are typical in the whole corpus (and their mutual information with words), one can construct high-quality word embedding with negligible errors. Mutual information between contexts and words can be encoded canonically as a sampling state, thus, Q-contexts can be fast constructed. Furthermore, we present an efficient and effective WEQ method, which is capable of extracting word embedding directly from these typical contexts. In practical scenarios, our algorithm runs 11 ~ 13 times faster than well-established methods. By comparing with well-known methods such as matrix factorization, word2vec, GloVe and fasttext, we demonstrate that our method achieves comparable performance on a variety of downstream NLP tasks, and in the meanwhile maintains run-time and resource advantages over all these baselines.

Classical algorithms in recommender systems (RS) mainly emphasis on achieving high accuracy and thus recommend items precisely matching a user's past choices. However, the user may gradually lose interest and crave something more inspiring. In psychology, curiosity is a critical human nature and can be efficient bootstrap exploratory behaviors, thus this phenomenon can be explained as insufficient stimulation to induce curiosity regard to recommended items. Inspired from the above, this work proposes a Curiosity-drive Recommendation Framework (CdRF) which incorporates a highly innovative Stimulus-evoked Curiosity mechanism (SeCM) together with a basic accuracy-oriented algorithm via Borda count. In SeCM, we first estimate the stimulus intensity appearing on each item for each user and then model personalized curiosity among the calculated intensities using Wundt curve. For the target user, the output of CdRF is a ranked list of N N items which are both relevant and highly curiousness. We conduct extensive experiments using four public datasets to evaluate the performance of each specification of SeCM as well as the whole framework CdRF. The results reveal that SeCM can flexibly generate diversified items and CdRF can increase diversity in terms of ILS, Newness and AD while compromising very little Precision. This kind of research also offers a way to understand both individual differences in curiosity and how curiosity contributes to item exploration at the level of RS.

We are pleased to present a special issue of Data Science and Engineering (DSE), which contains a collection of three extended papers from the APWeb-WAIM 2018 conference.Besides these three special issue papers, this DSE issue also has three regular research papers.APWeb-WAIM conferences focus on research, development, and applications in relation to Web information management, including a wide range of topics, such as text analysis, graph data processing, social networks, recommender systems, information retrieval, data streams, knowledge graph, data mining and application, query processing, machine learning, database and Web applications, big data, and blockchain.APWeb-WAIM 2018 was held in Macau during July 23-25, 2018, and attracted a total of 168 research paper submissions.The conference program committee selected 39 full research papers, 31 short papers, and six demonstration papers to be presented at the conference and published in the conference proceedings [1,2].The conference program also included keynote presentations by Prof.

The rapid growth of population has posed a challenge to people for discovering new followees in uni-directional social networks. Intuitively, a user's adoption of others as followees may motivated by her interest as well as social connection. Therefore, it is worth-while to consider both factors at the same time for better recommendations. Previous recommender works on implicit follow or not feedbacks become unqualified, mainly because of the coarse users' preferences inferring, which cannot distinguish whether one follows the other is based on her social connection or individual interest. In this paper, we present a new user recommendation method which is capable of recommending candidate followees who have similar interest and closer social connection relevant to a target user. As its core, a novel topic model namely UIS-LDA is designed to jointly model a user's preferences with respect to the set of latent interest topics and social topics. The experiments using Twitter dataset proves that our proposed method effective in improving the Precision, Conversion Rate F1 score and NDCG.

Collaborative filtering (CF) is an important and popular technology for recommender systems. However, current CF methods suffer from such problems as data sparsity, recommendation inaccuracy, and big-error in predictions. In this paper, we borrow ideas of object typicality from cognitive psychology and propose a novel typicality-based collaborative filtering recommendation method named TyCo. A distinct feature of typicality-based CF is that it finds "neighbors" of users based on user typicality degrees in user groups (instead of the corated items of users, or common users of items, as in traditional CF). To the best of our knowledge, there has been no prior work on investigating CF recommendation by combining object typicality. TyCo outperforms many CF recommendation methods on recommendation accuracy (in terms of MAE) with an improvement of at least 6.35 percent in Movielens data set, especially with sparse training data (9.89 percent improvement on MAE) and has lower time cost than other CF methods. Further, it can obtain more accurate predictions with less number of big-error predictions.

Recently, some systems allow users to rate and annotate resources, e.g., Movie Lens, and we consider that it provides a way to identify favor tags and annoying tags of a user by integrating user's rating and tags. In this paper, we reveal and elaborate on the limitations of current works on user profiling for personalized search in collaborative tagging systems. Then we propose a new multi-level user profiling model by integrating tags and ratings to achieve personalized search, which can reflect not only the user's favor but also a user's nuisances. To the best of our knowledge, this is the first effort to integrate the ratings and tags to model multi-level user profiles for personalized search.

In cognitive psychology, typicality refers to the degree of goodness of objects as exemplars in concepts. In this paper, we apply the idea of typicality analysis from cognitive psychology to query answering, and propose a novel method to answer typicality queries effectively based on theories in cognitive psychology. The proposed method adopts multi-prototype concept modeling and basic level category detection. By a systematic empirical evaluation using real data sets, we verify the accuracy and the effectiveness of our method on answering typicality queries.

With the increase of resource-sharing web sites such as YouTube1 and Flickr2, personalized search becomes more important and challenging, as users demand higher retrieval quality. To achieve this goal, personalized search needs to take users' personalized profiles and information needs into consideration. Collaborative tagging (also known as folksonomy [11]) systems allow users to annotate resources with their own tags, which provide a simple but powerful way for organizing, retrieving and sharing different types of social resources. In this paper, we examine the limitations of previous tag-based personalized search. To handle these limitations, we propose a new method to model user profiles and resource profiles in a collaborative tagging environment. A novel search method using such users' and resources' profiles is proposed to facilitate the desired personalization in resource search. We implement a prototype system named as FMRS. Experiments using FMRS data set and MovieLens data set show that our proposed method outperforms baseline methods.

Current recommendation methods are mainly classified into content-based, collaborative filtering and hybrid methods. These methods are based on similarity measurements among items or users. In this paper, we investigate recommendation systems from a new perspective based on object typicality and propose a novel typicality-based recommendation approach. Experiments show that our method outperforms compared methods on recommendation quality.