Image-based virtual try-on (VTON) aims to synthesize photorealistic images of a person wearing specified garments. Despite significant progress, building a universal VTON framework that can flexibly handle diverse and complex tasks remains a major challenge. Recent methods explore multi-task VTON frameworks guided by textual instructions, yet they still face two key limitations: (1) semantic gap between text instructions and reference images, and (2) data scarcity in complex scenarios. To address these challenges, we propose UniFit, a universal VTON framework driven by a Multimodal Large Language Model (MLLM). Specifically, we introduce an MLLM-Guided Semantic Alignment Module (MGSA), which integrates multimodal inputs using an MLLM and a set of learnable queries. By imposing a semantic alignment loss, MGSA captures cross-modal semantic relationships and provides coherent and explicit semantic guidance for the generative process, thereby reducing the semantic gap. Moreover, by devising a two-stage progressive training strategy with a self-synthesis pipeline, UniFit is able to learn complex tasks from limited data. Extensive experiments show that UniFit not only supports a wide range of VTON tasks, including multi-garment and model-to-model try-on, but also achieves state-of-the-art performance.

Songtao Jiang, Yuan Wang, Ruizhe Chen, Yan Zhang, Ruilin Luo, Bohan Lei, Yeying Jin, Sibo Song, ZhiBo Yang, Jimeng Sun, Jian Wu, Zuozhu Liu. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Recent advancements in large language models (LLMs) have given rise to the LLM-as-a-judge paradigm, showcasing their potential to deliver human-like judgments. However, in the field of machine translation (MT) evaluation, current LLM-as-a-judge methods fall short of learned automatic metrics. In this paper, we propose Multidimensional Multi-Agent Debate (M-MAD), a systematic LLM-based multi-agent framework for advanced LLM-as-a-judge MT evaluation. Our findings demonstrate that M-MAD achieves significant advancements by (1) decoupling heuristic MQM criteria into distinct evaluation dimensions for fine-grained assessments; (2) employing multi-agent debates to harness the collaborative reasoning capabilities of LLMs; (3) synthesizing dimension-specific results into a final evaluation judgment to ensure robust and reliable outcomes. Comprehensive experiments show that M-MAD not only outperforms all existing LLM-as-a-judge methods but also competes with state-of-the-art reference-based automatic metrics, even when powered by a suboptimal model like GPT-4o mini. Detailed ablations and analysis highlight the superiority of our framework design, offering a fresh perspective for LLM-as-a-judge paradigm. Our code and data are publicly available at https://github.com/SU-JIAYUAN/M-MAD.

Songtao Jiang, Yan Zhang, Yeying Jin, Zhihang Tang, Yangyang Wu, Yang Feng, Jian Wu, Zuozhu Liu. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

The strong capability of large language models (LLMs) has been applied to information extraction (IE) through either retrieval augmented prompting or instruction tuning (IT). However, the best way to incorporate information with LLMs for IE remains an open question. In this paper, we explore Retrieval Augmented Instruction Tuning (RA-IT) for IE, focusing on the task of open named entity recognition (NER). Specifically, for each training sample, we retrieve semantically similar examples from the training dataset as the context and prepend them to the input of the original instruction. To evaluate our RA-IT approach more thoroughly, we construct a Chinese IT dataset for open NER and evaluate RA-IT in both English and Chinese scenarios. Experimental results verify the effectiveness of RA-IT across various data sizes and in both English and Chinese scenarios. We also conduct thorough studies to explore the impacts of various retrieval strategies in the proposed RA-IT framework.

Multimodal large language models (MLLMs) have achieved remarkable success across various tasks. However, separate training of visual and textual encoders often results in a misalignment of the modality. Such misalignment may lead models to generate content that is absent from the input image, a phenomenon referred to as hallucination. These inaccuracies severely undermine the trustworthiness of MLLMs in real-world applications. Despite attempts to optimize text preferences to mitigate this issue, our initial investigation indicates that the trustworthiness of MLLMs remains inadequate. Specifically, these models tend to provide preferred answers even when the input image is heavily distorted. Analysis of visual token attention also indicates that the model focuses primarily on the surrounding context rather than the key object referenced in the question. These findings highlight a misalignment between the modalities, where answers inadequately leverage input images. Motivated by our findings, we propose Modality-Fair Preference Optimization (MFPO), which comprises three components: the construction of a multimodal preference dataset in which dispreferred images differ from originals solely in key regions; an image reward loss function encouraging the model to generate answers better aligned with the input images; and an easy-to-hard iterative alignment strategy to stabilize joint modality training. Extensive experiments on three trustworthiness benchmarks demonstrate that MFPO significantly enhances the trustworthiness of MLLMs. In particular, it enables the 7B models to attain trustworthiness levels on par with, or even surpass, those of the 13B, 34B, and larger models.

The goal of relationship classification (RC) is to predict the semantic relationship between two entities in a given sentence. With the advent of deep learning and pretrained language models, RC research has progressed by leaps and bounds. However, the current studies are focused mainly on predicting semantic relationships from a predefined set. How to recognize unseen relationships remains a challenge, which is also known as the zero-shot RC (ZSRC) task. Some ZSRC-related methods directly map relationship categories to numerical indices, constraining the model's ability to autonomously infer and understand these relationships, while others rely heavily on manual definitions. To address these issues and inspired by the way of reasoning in which humans perform RC tasks, we propose a new framework to handle the ZSRC task through inference on category attributes (ICAs). The main idea of ICA is to detect the semantic relationship between promises, which are RC sentences, and hypotheses, which are relational sentences of entities created by templates. Specifically, instead of manual design, we introduce two hypothesis templates derived from the label words (LWs) and descriptions (LDs) associated with each relationship. These templates are used to automatically convert the RC data into the textual entailment (TE) format. Furthermore, they are fine-tuned with a pretrained TE model, facilitating the acquisition of relational knowledge and enabling the generalization of semantic reasoning rules learned from seen classes to unseen classes. Moreover, to implement multirelationship semantic inference for all unseen classes, we propose an entailment difference mechanism to enhance the reasoning capability of the model. Besides the current ZSRC test setting, we also examine our method in an even more challenging setting to deal with data scarcity in real-world applications. The outstanding performance of ICA on the FewRel and Wiki-ZSL datasets demonstrates its effectiveness in the ZSRC task.

Event relations are crucial for narrative understanding and reasoning.Governed by nuanced logic, event relation extraction (ERE) is a challenging task that demands thorough semantic understanding and rigorous logical reasoning.In this paper, we conduct an in-depth investigation to systematically explore the capability of LLMs in understanding and applying event relation logic.More in detail, we first investigate the deficiencies of LLMs in logical reasoning across different tasks.Our study reveals that LLMs are not logically consistent reasoners, which results in their suboptimal performance on tasks that need rigorous reasoning.To address this, we explore three different approaches to endow LLMs with event relation logic, and thus enable them to generate more coherent answers across various scenarios.Based on our approach, we also contribute a synthesized dataset (LLM-ERL) involving high-order reasoning for evaluation and fine-tuning.Extensive quantitative and qualitative analyses on different tasks also validate the effectiveness of our approaches and provide insights for solving practical tasks with LLMs in future work.

Training or finetuning large-scale language models (LLMs) requires substantial computation resources, motivating recent efforts to explore parameter-efficient adaptation to downstream tasks. One approach is to treat these models as black boxes and use forward passes (Inference APIs) to interact with them. Current research focuses on adapting these black-box models to downstream tasks using gradient-free prompt optimization, but this often involves an expensive process of searching task-specific prompts. Therefore, we are motivated to study black-box language model adaptation without prompt search. Specifically, we introduce a label-enhanced cross-attention network called CrossTune, which models the semantic relatedness between the input text sequence and task-specific label descriptions. Its effectiveness is examined in the context of few-shot text classification. To improve the generalization of CrossTune, we utilize ChatGPT to generate additional training data through in-context learning. A switch mechanism is implemented to exclude low-quality ChatGPT-generated data. Through extensive experiments on seven benchmark text classification datasets, we demonstrate that our proposed approach outperforms the previous state-of-the-art gradient-free black-box tuning method by 5.7% on average. Even without using ChatGPT-augmented data, CrossTune performs better or comparably than previous black-box tuning methods, suggesting the effectiveness of our approach.

Causal reasoning is fundamental to human intelligence and crucial for effective decisionmaking in real-world environments.Despite recent advancements in large vision-language models (LVLMs), their ability to comprehend causality remains unclear.Previous work typically focuses on commonsense causality between events and/or actions, which is insufficient for applications like embodied agents and lacks the explicitly defined causal graphs required for formal causal reasoning.To overcome these limitations, we introduce a finegrained and unified definition of causality involving interactions between humans and/or objects.Building on the definition, we construct a novel dataset, CELLO, consisting of 14,094 causal questions across all four levels of causality: discovery, association, intervention, and counterfactual.This dataset surpasses traditional commonsense causality by including explicit causal graphs that detail the interactions between humans and objects.Extensive experiments on CELLO reveal that current LVLMs still struggle with causal reasoning tasks, but they can benefit significantly from our proposed CELLO-CoT, a causally inspired chain-of-thought prompting strategy.Both quantitative and qualitative analyses from this study provide valuable insights for future research.Our project page is at https: //github.com/OpenCausaLab/CELLO.

General-purpose Large Language Models (LLMs) like GPT-4 have achieved remarkable advancements in machine translation (MT) by leveraging extensive web content.On the other hand, translation-specific LLMs are built by pre-training on domain-specific monolingual corpora and fine-tuning with human-annotated translation data.Despite the superior performance, these methods either demand an unprecedented scale of computing and data or substantial human editing and annotation efforts.In this paper, we develop MT-Ladder, a novel model-agnostic and cost-effective tool to refine the performance of general LLMs for MT.MT-Ladder is trained on pseudo-refinement triplets which can be easily obtained from existing LLMs without additional human cost.During training, we propose a hierarchical finetuning strategy with an easy-to-hard schema, improving MT-Ladder's refining performance progressively.The trained MT-Ladder can be seamlessly integrated with any general-purpose LLMs to boost their translation performance.By utilizing Gemma-2B/7B as the backbone, MT-Ladder-2B can elevate raw translations to the level of top-tier open-source models (e.g., refining BigTranslate-13B with +6.91 BLEU and +3.52 COMET for XX→En), and MT-Ladder-7B can further enhance model performance to be on par with the state-of-theart GPT-4.Extensive ablation and analysis corroborate the effectiveness of MT-Ladder in diverse settings.Our code is available at https://github.com/fzp0424/MT-Ladder.

Large language models (LLMs) have demonstrated emergent capabilities across diverse reasoning tasks via popular Chains-of-Thought (COT) prompting.However, such a simple and fast COT approach often encounters limitations in dealing with complicated problems, while a thorough method, which considers multiple reasoning pathways and verifies each step carefully, results in slower inference.This paper addresses the challenge of enabling LLMs to autonomously select between fast and slow inference methods, thereby optimizing both efficiency and effectiveness.We introduce a dynamic decision-making framework that categorizes tasks into two distinct pathways: 'Fast', designated for tasks where the LLM quickly identifies a high-confidence solution, and 'Slow', allocated for tasks that the LLM perceives as complex and for which it has low confidence in immediate solutions as well as requiring more reasoning paths to verify.Experiments on five popular reasoning benchmarks demonstrated the superiority of the Dyna-Think over baselines.

Document-level Event Causality Identification (DECI) aims to recognize causal relations between events within a document.Recent studies focus on building a document-level graph for cross-sentence reasoning, but ignore important causal structures -there are one or two "central" events that prevail throughout the document, with most other events serving as either their cause or consequence.In this paper, we manually annotate central events for a systematical investigation and propose a novel DECI model, CHEER, which performs highorder reasoning while considering event centrality.First, we summarize a general GNNbased DECI model and provide a unified view for better understanding.Second, we design an Event Interaction Graph (EIG) involving the interactions among events (e.g., coreference) and event pairs, e.g., causal transitivity, cause(A, B) ∧ cause(B, C) ⇒ cause(A, C).Finally, we incorporate event centrality information into the EIG reasoning network via well-designed features and multi-task learning.We have conducted extensive experiments on two benchmark datasets.The results present great improvements (5.9% F1 gains on average) and demonstrate the effectiveness of each main component.

Large language models (LLMs) exhibited powerful capability in various natural language processing tasks.This work focuses on exploring LLM performance on zero-shot information extraction, with a focus on the ChatGPT and named entity recognition (NER) task.Inspired by the remarkable reasoning capability of LLM on symbolic and arithmetic reasoning, we adapt the prevalent reasoning methods to NER and propose reasoning strategies tailored for NER.First, we explore a decomposed question-answering paradigm by breaking down the NER task into simpler subproblems by labels.Second, we propose syntactic augmentation to stimulate the model's intermediate thinking in two ways: syntactic prompting, which encourages the model to analyze the syntactic structure itself, and tool augmentation, which provides the model with the syntactic information generated by a parsing tool.Besides, we adapt self-consistency to NER by proposing a two-stage majority voting strategy, which first votes for the most consistent mentions, then the most consistent types.The proposed methods achieve remarkable improvements for zero-shot NER across seven benchmarks, including Chinese and English datasets, and on both domainspecific and general-domain scenarios.In addition, we present a comprehensive analysis of the error types with suggestions for optimization directions.We also verify the effectiveness of the proposed methods on the few-shot setting and other LLMs. 1 * Corresponding authors. 1 Code available at: https://github.com/Emma1066/ Zero-Shot-NER-with-ChatGPT Input Text: The player who temporarily ranks second is German athlete Bao Lizzo, with a total score of 355.02 points, slightly lower than Lanwei.Gold Label: {"German":"Geo-Political Entity", "Lanwei": "Person", "BaoꞏLizzo": "Person"} Vanilla Ans: {"German athlete Bao Lizzo": "Person", "Lanwei": "Person"} TS-SC Ans: {"BaoꞏLizzo": "Person": "Person", "Lanwei": "Person", "German": "Geo-Political Entity"} ----------------------------------

Temporal knowledge graph (TKG) serves as an essential tool in modeling complex event relations among real-world entities. A temporal knowledge graph can be viewed as a collection of knowledge graph snapshots ordered by time. Reasoning over such graphs remains nontrivial as temporal causal dependencies between events are hard to capture. Current TKG reasoning methods only model pair-wise relations, which are limited in capturing higher-order dependencies between entities that are beyond dyadic connections. In this work, we aim to capture higher-order interactions of entities for TKG reasoning. To achieve this goal, we develop a Hierarchical Hypergraph Recurrent Attention Network on the type-induced entity hypergraph with multiple hierarchies to model the evolutionary pattern under different semantic granularities. The experimental analysis on benchmark datasets demonstrates the proposed model's superiority and elucidates the rationality of the hierarchical hypergraph modeling.

In dialog act recognition (DAR) of an utterance in a conversation, the prior studies have focused either on the global context using the whole utterances in the dialog, or the local context using the neighbouring utterance flow in the dialog. However, their methods attempt to deal with all types of dialogs indiscriminately. In this study, we propose a model to extract the local context information by an inter-utterance relation matching task (RMT), and a DAR framework to incorporate the local context information into a hierarchical network to fulfil both local and global context modeling. Extensive evaluations were conducted on a Mandarin dialog corpus and two benchmark English corpora. It is found that the different dialog types possess different window lengths for RMT, which is related to the length of subtopics in a given type of dialog. According to ablation experiments, the global information contributed more to the DAR in the hierarchical framework, while the contribution ratio of the local to the global context information was larger than 0.1. The results demonstrated that the proposed RMT and DAR framework significantly improved the DAR performance.

Software-Defined Vehicular Network (SDVN) is a networking architecture that can provide centralized control for vehicular networks. However, the design for routing policies in SDVNs is generally influenced by several limitations, such as frequent topological changes, complex service requests, and long model training time. Intelligent Digital Twin-based Software-Defined Vehicular Networks (IDT-SDVN) can overcome these weaknesses and maximize the advantages of the conventional SDVN architecture by enabling the controller to construct virtual network spaces and provide virtual instances of corresponding physical objects within the Digital Twin (DT). In this paper, we propose a junction-based hierarchical routing scheme in IDT-SDVN, namely, intelligent digital twin hierarchical (ELITE) routing. The proposed scheme is conducted in four phases: policy training and generation in the virtual network, and deployment and relay selection in physical networks. First, the policy learning phase employs several parallel agents in DT networks and derives multiple single-target policies. Second, the generation phase combines the learned policies and generates new policies based on complex communication requirements. Third, the deployment phase selects the most suitable generated policy according to the real-time network status and message types. A road path is calculated by the controller based on the selected policy and then sent to the requester vehicle. Finally, the relay selection phase is utilized to determine relay vehicles in a hop-by-hop process along the selected path. Simulation results demonstrate that ELITE achieves substantial improvements in terms of packet delivery ratio, end-to-end delay, and communication overhead compared with its counterparts.

Accurately delineating individual teeth and the gingiva in the three-dimension (3D) intraoral scanned (IOS) mesh data plays a pivotal role in many digital dental applications, e.g., orthodontics. Recent research shows that deep learning based methods can achieve promising results for 3D tooth segmentation, however, most of them rely on high-quality labeled dataset which is usually of small scales as annotating IOS meshes requires intensive human efforts. In this paper, we propose a novel self-supervised learning framework, named STSNet, to boost the performance of 3D tooth segmentation leveraging on large-scale unlabeled IOS data. The framework follows two-stage training, i.e., pre-training and fine-tuning. In pre-training, three hierarchical-level, i.e., point-level, region-level, cross-level, contrastive losses are proposed for unsupervised representation learning on a set of predefined matched points from different augmented views. The pretrained segmentation backbone is further fine-tuned in a supervised manner with a small number of labeled IOS meshes. With the same amount of annotated samples, our method can achieve an mIoU of 89.88%, significantly outperforming the supervised counterparts. The performance gain becomes more remarkable when only a small amount of labeled samples are available. Furthermore, STSNet can achieve better performance with only 40% of the annotated samples as compared to the fully supervised baselines. To the best of our knowledge, we present the first attempt of unsupervised pre-training for 3D tooth segmentation, demonstrating its strong potential in reducing human efforts for annotation and verification.

Traditionally, a debate usually requires a manual preparation process, including reading plenty of articles, selecting the claims, identifying the stances of the claims, seeking the evidence for the claims, etc.As the AI debate attracts more attention these years, it is worth exploring the methods to automate the tedious process involved in the debating system.In this work, we introduce a comprehensive and large dataset named IAM, which can be applied to a series of argument mining tasks, including claim extraction, stance classification, evidence extraction, etc.Our dataset is collected from over 1k articles related to 123 topics.Near 70k sentences in the dataset are fully annotated based on their argument properties (e.g., claims, stances, evidence, etc.).We further propose two new integrated argument mining tasks associated with the debate preparation process: (1) claim extraction with stance classification (CESC) and (2) claim-evidence pair extraction (CEPE).We adopt a pipeline approach and an end-to-end method for each integrated task separately.Promising experimental results are reported to show the values and challenges of our proposed tasks, and motivate future research on argument mining. 1

Document-level Event Causality Identification (DECI) aims to identify event-event causal relations in a document. Existing works usually build an event graph for global reasoning across multiple sentences. However, the edges between events have to be carefully designed through heuristic rules or external tools. In this paper, we propose a novel Event Relational Graph TransfOrmer (ERGO) framework for DECI, to ease the graph construction and improve it over the noisy edge issue. Different from conventional event graphs, we define a pair of events as a node and build a complete event relational graph without any prior knowledge or tools. This naturally formulates DECI as a node classification problem, and thus we capture the causation transitivity among event pairs via a graph transformer. Furthermore, we design a criss-cross constraint and an adaptive focal loss for the imbalanced classification, to alleviate the issues of false positives and false negatives. Extensive experiments on two benchmark datasets show that ERGO greatly outperforms previous state-of-the-art (SOTA) methods (12.8% F1 gains on average).

Most sentence embedding techniques heavily rely on expensive human-annotated sentence pairs as the supervised signals.Despite the use of large-scale unlabeled data, the performance of unsupervised methods typically lags far behind that of the supervised counterparts in most downstream tasks.In this work, we propose a semi-supervised sentence embedding framework, GenSE, that effectively leverages large-scale unlabeled data.Our method include three parts: 1) Generate: A generator/discriminator model is jointly trained to synthesize sentence pairs from open-domain unlabeled corpus; 2) Discriminate: Noisy sentence pairs are filtered out by the discriminator to acquire high-quality positive and negative sentence pairs; 3) Contrast: A prompt-based contrastive approach is presented for sentence representation learning with both annotated and synthesized data.Comprehensive experiments show that GenSE achieves an average correlation score of 85.19 on the STS datasets and consistent performance improvement on four domain adaptation tasks, significantly surpassing the state-of-the-art methods and convincingly corroborating its effectiveness and generalization ability. 1

Dialogue summarization is abstractive in nature, making it suffer from factual errors.The factual correctness of summaries has the highest priority before practical applications.Many efforts have been made to improve faithfulness in text summarization.However, there is a lack of systematic study on dialogue summarization systems.In this work, we first perform the fine-grained human analysis on the faithfulness of dialogue summaries and observe that over 35% of generated summaries are faithfully inconsistent respective the source dialogues.Furthermore, we present a new model-level faithfulness evaluation method.It examines generation models with multi-choice questions created by rule-based transformations.Experimental results show that our evaluation schema is a strong proxy for the factual correctness of summarization models.The humanannotated faithfulness samples and the evaluation toolkit are released to facilitate future research toward faithful dialogue summarization.Code

Deep latent variable models such as variational autoencoders and energy-based models are widely used for neural text generation. Most of them focus on matching the prior distribution with the posterior distribution of the latent variable for text reconstruction. In addition to instance-level reconstruction, this paper aims to integrate contrastive learning in the latent space, forcing the latent variables to learn high-level semantics by exploring inter-instance relationships. Experiments on various text generation benchmarks show the effectiveness of our proposed method. We also empirically show that our method can mitigate the posterior collapse issue for latent variable based text generation models.

Session-based recommendation aims to predict a user's next action based on previous actions in the current session. The major challenge is to capture authentic and complete user preferences in the entire session. Recent work utilizes graph structure to represent the entire session and adopts Graph Neural Network (GNN) to encode session information. This modeling choice has been proved to be effective and achieved remarkable results. However, most of the existing studies only consider each item within the session independently and do not capture session semantics from a high-level perspective. Such limitation often leads to severe information loss and increases the difficulty of capturing long-range dependencies within a session. Intuitively, compared with individual items, a session snippet, i.e., a group of locally consecutive items, is able to provide supplemental user intents which are hardly captured by existing methods. In this work, we propose to learn multi-granularity consecutive user intent unit to improve the recommendation performance. Specifically, we creatively propose Multi-granularity Intent Heterogeneous Session Graph (MIHSG) which captures the interactions between different granularity intent units and relieves the burden of long-dependency. Moreover, we propose the Intent Fusion Ranking (IFR) module to compose the recommendation results from various granularity user intents. Compared with current methods that only leverage intents from individual items, IFR benefits from different granularity user intents to generate more accurate and comprehensive session representation, thus eventually boosting recommendation performance. We conduct extensive experiments on five session-based recommendation datasets and the results demonstrate the effectiveness of our method. Compared to current state-of-the-art methods, we achieve as large as 10.21% gain on [email protected] and 15.53% gain on [email protected]

Yan Zhang, Ruidan He, Zuozhu Liu, Lidong Bing, Haizhou Li. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

Chen Zhang, Yiming Chen, Luis Fernando D’Haro, Yan Zhang, Thomas Friedrichs, Grandee Lee, Haizhou Li. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

As unlabeled data carry rich task-relevant information, they are proven useful for fewshot learning of language model.The question is how to effectively make use of such data.In this work, we revisit the self-training technique for language model fine-tuning and present a state-of-the-art prompt-based fewshot learner, SFLM.Given two views of a text sample via weak and strong augmentation techniques, SFLM generates a pseudo label on the weakly augmented version.Then, the model predicts the same pseudo label when fine-tuned with the strongly augmented version.This simple approach is shown to outperform other state-of-the-art supervised and semi-supervised counterparts on six sentence classification and six sentence-pair classification benchmarking tasks.In addition, SFLM only relies on a few in-domain unlabeled data.We conduct a comprehensive analysis to demonstrate the robustness of our proposed approach under various settings, including augmentation techniques, model scale, and fewshot knowledge transfer across tasks.

Link prediction is the task of predicting missing connections between entities in the knowledge graph (KG). While various forms of models are proposed for the link prediction task, most of them are designed based on a few known relation patterns in several well-known datasets. Due to the diversity and complexity nature of the real-world KGs, it is inherently difficult to design a model that fits all datasets well. To address this issue, previous work has tried to use Automated Machine Learning (AutoML) to search for the best model for a given dataset. However, their search space is limited only to bilinear model families. In this paper, we propose a novel Neural Architecture Search (NAS) framework for the link prediction task. First, the embeddings of the input triplet are refined by the Representation Search Module. Then, the prediction score is searched within the Score Function Search Module. This framework entails a more general search space, which enables us to take advantage of several mainstream model families, and thus it can potentially achieve better performance. We relax the search space to be continuous so that the architecture can be optimized efficiently using gradient-based search strategies. Experimental results on several benchmark datasets demonstrate the effectiveness of our method compared with several state-of-the-art approaches.

Previous works on knowledge-to-text generation take as input a few RDF triples or keyvalue pairs conveying the knowledge of some entities to generate a natural language description.Existing datasets, such as WIKIBIO, WebNLG, and E2E, basically have a good alignment between an input triple/pair set and its output text.However, in practice, the input knowledge could be more than enough, since the output description may only cover the most significant knowledge.In this paper, we introduce a large-scale and challenging dataset to facilitate the study of such a practical scenario in KG-to-text.Our dataset involves retrieving abundant knowledge of various types of main entities from a large knowledge graph (KG), which makes the current graph-to-sequence models severely suffer from the problems of information loss and parameter explosion while generating the descriptions.We address these challenges by proposing a multi-graph structure that is able to represent the original graph information more comprehensively.Furthermore, we also incorporate aggregation methods that learn to extract the rich graph information.Extensive experiments demonstrate the effectiveness of our model architecture.1

Graph embedding (GE) methods embed nodes (and/or edges) in graph into a low-dimensional semantic space, and have shown its effectiveness in modeling multi-relational data.However, existing GE models are not practical in real-world applications since it overlooked the streaming nature of incoming data.To address this issue, we study the problem of continual graph representation learning which aims to continually train a graph embedding model on new data to learn incessantly emerging multi-relational data while avoiding catastrophically forgetting old learned knowledge.Moreover, we propose a disentangle-based continual graph representation learning (DiC-GRL) framework inspired by the human's ability to learn procedural knowledge.The experimental results show that DiCGRL could effectively alleviate the catastrophic forgetting problem and outperform state-of-the-art continual learning models.* This work is done when Xiaoyu Kou was interning at Pattern Recognition Center, WeChat AI, Tencent Inc, China !"#"$% &'"(" )*$+,--, &'"(" )"-*" .//&'"(" .//,01/+"( 2#,3*4,/5 6+, 7/*5,4 85"5,3 85"5, 9: ;"<"** ="<>,# ?#"3,# @.B9'*/39/ ?*#35 ="4> 9: 78 @+*$"C9

AMR-to-text generation is used to transduce Abstract Meaning Representation structures (AMR) into text.A key challenge in this task is to efficiently learn effective graph representations.Previously, Graph Convolution Networks (GCNs) were used to encode input AMRs, however, vanilla GCNs are not able to capture non-local information and additionally, they follow a local (first-order) information aggregation scheme.To account for these issues, larger and deeper GCN models are required to capture more complex interactions.In this paper, we introduce a dynamic fusion mechanism, proposing Lightweight Dynamic Graph Convolutional Networks (LDGCNs) that capture richer non-local interactions by synthesizing higher order information from the input graphs.We further develop two novel parameter saving strategies based on the group graph convolutions and weight tied convolutions to reduce memory usage and model complexity.With the help of these strategies, we are able to train a model with fewer parameters while maintaining the model capacity.Experiments demonstrate that LDGCNs outperform stateof-the-art models on two benchmark datasets for AMR-to-text generation with significantly fewer parameters.

BERT is inefficient for sentence-pair tasks such as clustering or semantic search as it needs to evaluate combinatorially many sentence pairs which is very time-consuming.Sentence BERT (SBERT) attempted to solve this challenge by learning semantically meaningful representations of single sentences, such that similarity comparison can be easily accessed.However, SBERT is trained on corpus with high-quality labeled sentence pairs, which limits its application to tasks where labeled data is extremely scarce.In this paper, we propose a lightweight extension on top of BERT and a novel self-supervised learning objective based on mutual information maximization strategies to derive meaningful sentence embeddings in an unsupervised manner.Unlike SBERT, our method is not restricted by the availability of labeled data, such that it can be applied on different domain-specific corpus.Experimental results show that the proposed method significantly outperforms other unsupervised sentence embedding baselines on common semantic textual similarity (STS) tasks and downstream supervised tasks.It also outperforms SBERT in a setting where in-domain labeled data is not available, and achieves performance competitive with supervised methods on various tasks.

Recently, the embedding-based recommendation models (e.g., matrix factorization and deep models) have been prevalent in both academia and industry due to their effectiveness and flexibility. However, they also have such intrinsic limitations as lacking explainability and suffering from data sparsity. In this paper, we propose an end-to-end joint learning framework to get around these limitations without introducing any extra overhead by distilling structured knowledge from a differentiable path-based recommendation model. Through extensive experiments, we show that our proposed framework can achieve state-of-the-art recommendation performance and meanwhile provide interpretable recommendation reasons.

Neural networks have been extensively used in recommender systems. Embedding layers are not only necessary but also crucial for neural models in recommendation as a typical discrete task. In this article, we argue that the widely used l 2 regularization for normal neural layers (e.g., fully connected layers) is not ideal for embedding layers from the perspective of regularization theory in Reproducing Kernel Hilbert Space. More specifically, the l 2 regularization corresponds to the inner product and the distance in the Euclidean space where correlations between discrete objects (e.g., items) are not well captured. Inspired by this observation, we propose a graph-based regularization approach to serve as a counterpart of the l 2 regularization for embedding layers. The proposed regularization incurs almost no extra computational overhead especially when being trained with mini-batches. We also discuss its relationships to other approaches (namely, data augmentation, graph convolution, and joint learning) theoretically. We conducted extensive experiments on five publicly available datasets from various domains with two state-of-the-art recommendation models. Results show that given a kNN (k-nearest neighbor) graph constructed directly from training data without external information, the proposed approach significantly outperforms the l 2 regularization on all the datasets and achieves more notable improvements for long-tail users and items.

Dependency trees convey rich structural information that is proven useful for extracting relations among entities in text.However, how to effectively make use of relevant information while ignoring irrelevant information from the dependency trees remains a challenging research question.Existing approaches employing rule based hard-pruning strategies for selecting relevant partial dependency structures may not always yield optimal results.In this work, we propose Attention Guided Graph Convolutional Networks (AGGCNs), a novel model which directly takes full dependency trees as inputs.Our model can be understood as a soft-pruning approach that automatically learns how to selectively attend to the relevant sub-structures useful for the relation extraction task.Extensive results on various tasks including cross-sentence n-ary relation extraction and large-scale sentence-level relation extraction show that our model is able to better leverage the structural information of the full dependency trees, giving significantly better results than previous approaches.

Almost all the existing network embedding methods learn to map the node IDs to their corresponding node embeddings. This design principle, however, hinders the existing methods from being applied in real cases. Node ID is not generalizable and, thus, the existing methods have to pay great effort in cold-start problem. The heterogeneous network usually requires extra work to encode node types, as node type is not able to be identified by node ID. Node ID carries rare information, resulting in the criticism that the existing methods are not robust to noise. To address this issue, we introduce Compositional Network Embedding, a general inductive network representation learning framework that generates node embeddings by combining node features based on the "principle of compositionally". Instead of directly optimizing an embedding lookup based on arbitrary node IDs, we learn a composition function that infers node embeddings by combining the corresponding node attribute embeddings through a graph-based loss. For evaluation, we conduct the experiments on link prediction under three different settings. The results verified the effectiveness and generalization ability of compositional network embeddings, especially on unseen nodes.

Recently, neural models for information retrieval are becoming increasingly popular. They provide effective approaches for product search due to their competitive advantages in semantic matching. However, it is challenging to use graph-based features, though proved very useful in IR literature, in these neural approaches. In this paper, we leverage the recent advances in graph embedding techniques to enable neural retrieval models to exploit graph-structured data for automatic feature extraction. The proposed approach can not only help to overcome the long-tail problem of click-through data, but also incorporate external heterogeneous information to improve search results. Extensive experiments on a real-world e-commerce dataset demonstrate significant improvement achieved by our proposed approach over multiple strong baselines both as an individual retrieval model and as a feature used in learning-to-rank frameworks.

We focus on graph-to-sequence learning, which can be framed as transducing graph structures to sequences for text generation. To capture structural information associated with graphs, we investigate the problem of encoding graphs using graph convolutional networks (GCNs). Unlike various existing approaches where shallow architectures were used for capturing local structural information only, we introduce a dense connection strategy, proposing a novel Densely Connected Graph Convolutional Network (DCGCN). Such a deep architecture is able to integrate both local and non-local features to learn a better structural representation of a graph. Our model outperforms the state-of-the-art neural models significantly on AMR-to-text generation and syntax-based neural machine translation.

This paper studies the problem of social network embedding without relying on network structures that are usually not observed in many cases. We address that the information diffusion process across networks naturally reflects rich proximity relationships between users. Meanwhile, social networks contain multiple communities regularizing communication pathways for information propagation. Based on the above observations, we propose a probabilistic generative model, called COSINE, to learn community-preserving social network embeddings from the recurrent and time-stamped social contagion logs, namely information diffusion cascades. The learned embeddings therefore capture the high-order user proximities in social networks. Leveraging COSINE, we are able to discover underlying social communities and predict temporal dynamics of social contagion. Experimental results on both synthetic and real-world datasets show that our proposed model significantly outperforms the existing approaches.

Neural network techniques are widely used in network embedding, boosting the result of node classification, link prediction, visualization and other tasks in both aspects of efficiency and quality. All the state of art algorithms put effort on the neighborhood information and try to make full use of it. However, it is hard to recognize core periphery structures simply based on neighborhood. In this paper, we first discuss the influence brought by random-walk based sampling strategies to the embedding results. Theoretical and experimental evidences show that random-walk based sampling strategies fail to fully capture structural equivalence. We present a new method, SNS, that performs network embeddings using structural information (namely graphlets) to enhance its quality. SNS effectively utilizes both neighbor information and local-subgraphs similarity to learn node embeddings. This is the first framework that combines these two aspects as far as we know, positively merging two important areas in graph mining and machine learning. Moreover, we investigate what kinds of local-subgraph features matter the most on the node classification task, which enables us to further improve the embedding quality. Experiments show that our algorithm outperforms other unsupervised and semi-supervised neural network embedding algorithms on several real-world datasets.

Real-time location inference of social media users is the fundamental of some spatial applications such as localized search and event detection. While tweet text is the most commonly used feature in location estimation, most of the prior works suffer from either the noise or the sparsity of textual features. In this paper, we aim to tackle these two problems. We use topic modeling as a building block to characterize the geographic topic variation and lexical variation so that "one-hot" encoding vectors will no longer be directly used. We also incorporate other features which can be extracted through the Twitter streaming API to overcome the noise problem. Experimental results show that our RATE algorithm outperforms several benchmark methods, both in the precision of region classification and the mean distance error of latitude and longitude regression.

It is a great challenge to model and mine the e-commercial data, which is made up of multiple types of objects, such as products, users, comments and tags. To model the complicated interactive relationships in the the e-commercial data, we propose to transform the complex e-commercial data into a text-rich heterogeneous e-commercial network. Then three neural network based embedding algorithms named WTL (Weighted Text Learning), IBL (Identity Based Learning) and IBTSL (Identity Based Two Steps Learning) are proposed to consider both the network structure information and heterogeneous nodes attributes identity information to learn the embeddings. The key idea of our models is to map all objects in the e-commercial network to a same low-dimensional vector space, which is useful to produce meaningful features for many applications such as product classification, comment classification, product attributes forecasting, recommendation, and so on. Our algorithms are compared with other existing advanced methods on a real large-scale e-commercial dataset. Several applications are set to evaluate the effectivity of the learned embeddings. The experimental results show that the embeddings generated by our algorithms have superior performance in each application.

Recently, online social networks are becoming increasingly popular platforms for social interactions. Understanding how information propagates in such networks is important for personalization and recommendation in social search.

Influence maximization, the fundamental of viral marketing, aims to find top-$K$ seed nodes maximizing influence spread under certain spreading models. In this paper, we study influence maximization from a game perspective. We propose a Coordination Game model, in which every individuals make their decisions based on the benefit of coordination with their network neighbors, to study information propagation. Our model serves as the generalization of some existing models, such as Majority Vote model and Linear Threshold model. Under the generalized model, we study the hardness of influence maximization and the approximation guarantee of the greedy algorithm. We also combine several strategies to accelerate the algorithm. Experimental results show that after the acceleration, our algorithm significantly outperforms other heuristics, and it is three orders of magnitude faster than the original greedy method.

Network embedding is a classical task which aims to map the nodes of a network to low-dimensional vectors. Most of the previous network embedding methods are trained in an unsupervised scheme. Then the learned node embeddings can be used as inputs of many machine learning tasks such as node classification, attribute inference. However, the discrimination validity of the node embeddings maybe improved by considering the node label information and the node attribute information. Inspired by traditional semi-supervised learning techniques, we explore to train the node embeddings and the node classifiers simultaneously with the text attributes information in a flexible framework. We present NLSTNE (Non-Linear Smoothed Transductive Network Embedding), a transductive network embedding method, whose embeddings are enhanced by modeling the non-linear pairwise similarity between the nodes and the non-linear relationship between the nodes and the text attributes. We use the node classification task to evaluate the quality of node embeddings learned by different models on four real-world network datasets . The experimental results demonstrate that our model outperforms several state-of-the-art network embedding methods.

Community detection is a hot topic for researchers in the fields including graph theory, social networks and biological networks. Generally speaking, a community refers to a group of densely linked nodes in the network. Nodes usually have more than one community label, indicating their multiple roles or functions in the network. Unfortunately, existing solutions aiming at overlapping-community-detection are not capable of scaling to large-scale networks with millions of nodes and edges. In this paper, we propose a fast overlapping-communitydetection algorithm - FOX. In the experiment on a network with 3.9 millions nodes and 20 millions edges, the detection finishes in 14 minutes and provides the most qualified results. The second fastest algorithm, however, takes ten times longer to run. As for another network with 22 millions nodes and 127 millions edges, our algorithm is the only one that can provide an overlapping community detection result and it only takes 238 minutes. Our algorithm draws lessons from potential games, a concept in game theory. We measure the closeness of a node to a community by counting the number of triangles formed by the node and two other nodes form the community. Potential games ensure that the algorithm can reach convergence. We also extend the exploitation of triangle to open-triangle, which enlarges the scale of the detected communities.

Similarity measurement is fundamental to many data mining and information retrieval tasks such as link prediction and relevance-based search. Conventional similarity measurement relies more on homogenous linkage relation and content information. However, these measurements cannot take full advantage of the data structure as heterogenous graph gains increasing popularity. Moreover, the scalability of these methods also faces challenge with the never-ending growth of big data in real world. In this paper, we propose a new similarity measurement called HBGSim based on the heterogeneous structured data. HBGSim combines both local and global features by a two-stage process. We make a comparison between our measurement and some traditional methods on DBLP1dataset for evaluation and the experimental results show that our method outperforms the others.

Knowledge graphs are recently used for enriching query representations in an entity-aware way for the rich facts organized around entities in it.However, few of the methods pay attention to non-entity words and clicked websites in queries, which also help conveying user intent.In this paper, we tackle the problem of intent understanding with innovatively representing entity words, refiners and clicked urls as intent topics in a unified knowledge graph based framework, in a way to exploit and expand knowledge graph which we call 'tailor'.We collaboratively exploit global knowledge in knowledge graphs and local contexts in query log to initialize intent representation, then propagate the enriched features in a graph consisting of intent topics using an unsupervised algorithm.The experiments prove intent topics with knowledge graph enriched features significantly enhance intent understanding.

The articles in this special issue address the technologies and applications supported by the Internet of vehicles (IoV). The new IoT is driving the evolution of conventional vehicle networks into the IoV. The difference of the vehicle concept in VANET and IoV makes these two scenarios essentially different in the device, communications, networking, and services aspects. In VANET, a vehicle is mainly considered as a node to disseminate messages among vehicles. In the IoV paradigm, each vehicle is considered as a smart object equipped with a powerful multisensor platform, communications technologies, computation units, and Internet protocol (IP)-based connectivity to the Internet and to other vehicles either directly or indirectly. In addition, a vehicle in IoV is envisioned as a multicommunication model, enabling the interactions between intravehicle components, vehicles and vehicles, vehicles and road, and vehicles and people.

In this paper, we investigate the problem of making better use of semantic knowledge obtained from different encyclopedia sources. We propose a framework to integrate different encyclopedias and reorganize the information. We also utilize Learning to Rank models to distill out more functional knowledge from the encyclopedic information and then align the knowledge with a WordNet-like ontology. Finally as a demonstration, a Chinese semantic knowledge repository named JNet is constructed based on this framework. Experiments show that the proposed methods work well and the three steps reinforce each other towards a more powerful ontology.

The rapid development of Web2.0 leads to significant information redundancy.Especially for a complex news event, it is difficult to understand its general idea within a single coherent picture.A complex event often contains branches, intertwining narratives and side news which are all called storylines.In this paper, we propose a novel solution to tackle the challenging problem of storylines extraction and reconstruction.Specifically, we first investigate two requisite properties of an ideal storyline.Then a unified algorithm is devised to extract all effective storylines by optimizing these properties at the same time.Finally, we reconstruct all extracted lines and generate the high-quality story map.Experiments on real-world datasets show that our method is quite efficient and highly competitive, which can bring about quicker, clearer and deeper comprehension to readers.

Summarization methods such as timelines have greatly helped people to understand all kinds of news events within limited time. However, there are few studies probing into cross-media summarization, for example, generating timelines which contain both texts and images that can reinforce each other. In this paper, we tackle this important and challenging problem by proposing a novel solution. Specifically, we first reveal three requisite characteristics of an ideal image-text timeline. With the idea of recommendation, all these requisites will be modeled respectively, and fused compactly in a unified cross-media framework. Finally, we put all sentences and images into either the schema of referrer or the schema of recommended candidate, and the former recommends the latter. After changing their roles iteratively, we can achieve the optimal timelines which will significantly improve user experience and satisfaction. Experiments on real-world datasets show that the timelines generated by our framework outperform several competitive baselines.

In many cases, people would like to read the news with great importance on the Internet. However, what users can grasp covers a very small part compared with the huge amount of news which never stops increasing. In this paper, we try to find what users are most likely to be interested in. We notice that media focus plays an essential role in distinguishing news topics and user attention is also an important factor. Therefore, we first propose five strategies which only exploit media focus to decide news influence impact. Then we provide three strategies to combine user attention with media focus. Meanwhile, we also take four types of interaction between user attention and media focus into consideration. To the best of our knowledge, this is the first work to establish different models for computing influence decay of news topics. Experiments show that better influence scores will be achieved by a decay algorithm based on Ebbinghaus forgetting curve and information fusion by considering interactions between user attention and media focus.

Under the joint influence of the presentation of search results and users' browsing and clicking habits, the click probability distribution does not merely obey a monotonic decreasing Zipf function. In this paper, we present evidence that the click behavior on the entries of search engines' result pages is influenced by Serial Position Effect, which is independent of how these entries are ordered, and introduce a new function to characterize the click probability distribution.

Manual timelines have greatly helped us to keep pace with the big world. In this paper, we introduce a novel solution which generates image-text timelines for news events based on Evolutionary Image-Text Summarization, which is an important and challenging problem. We first extract image's semantic information under translation model, and then fuse the high quality images with text timeline under an image assignment algorithm which can optimize the global coordination of the final timeline. The experimental results show that news readers can receive more satisfaction from the image-text timelines we generate.

As a well-known semantic repository, WordNet is widely used in many applications. However, due to costly edit and maintenance, WordNet's capability of keeping up with the emergence of new concepts is poor compared with on-line encyclopedias such as Wikipedia. To keep WordNet current with folk wisdom, we propose a method to enhance WordNet automatically by merging Wikipedia entities into WordNet, and construct an enriched ontology, named as WorkiNet. WorkiNet keeps the desirable structure of WordNet. At the same time, it captures abundant information from Wikipedia. We also propose a learning approach which is able to generate a tailor-made semantic concept collection for a given document collection. The learning process takes the characteristics of the given document collection into consideration and the semantic concepts in the tailor-made collection can be used as new features for document representation. The experimental results show that the adaptively generated feature space can outperform a static one significantly in text mining tasks, and WorkiNet dominates WordNet most of the time due to its high coverage.

Classic news summarization plays an important role with the exponential document growth on the Web. Many approaches are proposed to generate summaries but seldom simultaneously consider evolutionary characteristics of news plus to traditional summary elements. Therefore, we present a novel framework for the web mining problem named Evolutionary Timeline Summarization (ETS). Given the massive collection of time-stamped web documents related to a general news query, ETS aims to return the evolution trajectory along the timeline, consisting of individual but correlated summaries of each date, emphasizing relevance, coverage, coherence and cross-date diversity. ETS greatly facilitates fast news browsing and knowledge comprehension and hence is a necessity. We formally formulate the task as an optimization problem via iterative substitution from a set of sentences to a subset of sentences that satisfies the above requirements, balancing coherence/diversity measurement and local/global summary quality. The optimized substitution is iteratively conducted by incorporating several constraints until convergence. We develop experimental systems to evaluate on 6 instinctively different datasets which amount to 10251 documents. Performance comparisons between different system-generated timelines and manually created ones by human editors demonstrate the effectiveness of our proposed framework in terms of ROUGE metrics.

It is well known that synonymous and polysemous terms often bring in some noises when calculating the similarity between documents. Existing ontology-based document representation methods are static, hence, the chosen semantic concept set for representing a document has a fixed resolution and it is not adaptable to the characteristics of a document collection and the text mining problem in hand. We propose an Adaptive Concept Resolution (ACR) model to overcome this issue. ACR can learn a concept border from an ontology taking into consideration of the characteristics of a particular document collection. Then this border can provide a tailor-made semantic concept representation for a document coming from the same domain. Another advantage of ACR is that it is applicable in both classification task where the groups are given in the training document set, and clustering task where no group information is available. Furthermore, the result of this model is not sensitive to the model parameter. The experimental results show that ACR outperforms an existing static method significantly.

Welcome to this special issue of the Wiley's Wireless Communications and Mobile Computing Journal. This special issue is devoted to the topic of the latest research and development in the field of wireless communications and networking. With the explosive growth of the ever-increasing users' demands for broadband services, there are a number of emerging wireless technologies, including cognitive radio, wireless sensor networks, WiMAX/LTE, adaptive communications, etc. These technologies have led to significant innovations that enable systems in providing more bandwidth with convenient and inexpensive deployment and mobility. This issue called for papers in various aspects of recent wireless communications and mobile networking. In this special issue, we selected 11 papers to show the recent advances. The papers cover both topical and innovative areas. A detailed overview of the selected works is given below. In 'A Software-Defined Radio Based Cognitive Radio Demonstration over FM Band', R. Zhou et al. present a software-defined radio (SDR) based cognitive radio (CR) implementation and demonstration over the frequency. The authors have proposed, implemented, and demonstrated a frequency hopping scheme over multiple spectrum holes to support multiple secondary users while attaining the minimum interference among the users. In 'A Sensing-Based Cognitive Coexistence Method for Interfering Infrastructure and Ad-Hoc Systems', S. Geirhofer et al. present a novel cognitive coexistence framework, which enables an infrastructure system to reduce interference to ad hoc or peer-to-peer communication links in close proximity. They study how the centralized resource allocation can accommodate the ad hoc links based on sensing and predicting their interference patterns. Results illustrate that utilizing the superior flexibility of the infrastructure links can effectively mitigate interference. In 'Medium Access Control Protocols in Cognitive Radio Networks', J. Xiang et al. make a comprehensive survey of the state-of-the-art MAC and categorize the MAC protocols on the basis of spectrum sharing mechanisms, i.e., overlay and underlay modes. In each mode, they discuss the protocols in both centralized and distributed manners. Finally, they summarize the schemes and identify several open research issues in the realization of cost-efficient MAC protocols. In 'Constructing Secured Cognitive Wireless Networks: Experiences and Challenges', C. Li et al. conduct a high-level survey to review and reflect the state-of-the-art work on the security issues in cognitive radio networks. They focus on analyzing the security system at the macroscopic level, where both protection and detection are considered to be the most essential parts to ensure security in the whole network. Furthermore, they investigate special characteristics of cognitive radio network at different protocol layers, including physical layer, link layer, network layer, transport layer, and application layer. In 'Adaptive Antenna Selection at Mobile Stations for SDMA in WiMAX Networks', Wang et al. present a new antenna selection protocol associated with low signaling and implementation complexity overhead. The protocol has been developed for next-generation IEEE 802.16 mobile stations operating in space division multiple access (SDMA) mode. The authors demonstrate through simulation and analysis that the proposed protocol performs better than traditional IEEE 802.16 terminals that do not have support for antenna selection. The robustness of the protocol is further demonstrated by its capability to adapt different channel conditions. In 'Per-User Service Model for Opportunistic Scheduling Scheme over Fading Channels', Dianati et al. investigated opportunistic scheduling for data transmissions from a single base station to several users. They demonstrate that the received service by a single user can be modeled analytically as a finite-state Markov process under saturated conditions when user queues are heavily loaded. The simulation results obtained closely match those predicted by the proposed analytical model demonstrating the efficacy and accuracy of the model. In 'TinyNET—A Tiny Network framework for TinyOS: Description, Implementation and Experimentation', Castellani et al. present the architecture of an interesting modular network framework that is aimed at speeding up the development and integration of wireless sensor network applications and protocols in TinyOS. By exploiting the framework interfaces, components can be rapidly developed, debugging is made much easier, and software maintenance is simplified because of the in-built modularity. The authors also present a proof-of-concept scenario that explores the proposed framework's basic functionalities and the associated overheads it incurs. The results demonstrate that TinyNET provides significant benefits with no additional complexity and memory costs. In 'Towards Utility-optimal Random Access without Message Passing', Y. Yi et al. study a generalization of CAMA algorithm. In the continuous-time model, a proof is presented of the convergence of these adaptive CSMA algorithms to be arbitrarily close to utility optimality, without assuming that the network dynamics converge to an equilibrium in between consecutive CSMA parameter updates. In the more realistic, slotted-time model, the impact of collisions on the utility achieved is characterized, and the tradeoff between optimality and short-term fairness is quantified. In 'Cross-layer design of joint routing and rate control in ad hoc wireless networks', Ju-Lan et al. present an adaptive modulation coding scheme in multi-hop wireless network in which each node independently selects its cross-layer parameter vector for each packet that it forwards. Furthermore, to enable throughput-effective network operations, a cross-layer scheme is presented under which each node configures its parameter vector by using the link transport capacity measure that it computes as a key metric. The results demonstrate the efficiency of the proposed schemes employing the link transport capacity measure. In 'PC-MAC: Pico Cellular MAC Protocol for Motorway Vehicular Multimedia Communication (extended version)', Bilal et al. presented a detailed analysis of real vehicular traffic data taken from inductive loops on a motorway, and developed their own vehicular simulator to study real-time vehicular traffic scenarios in which a main BS is responsible for multimedia communication. The pico cellular medium access control (PC-MAC) protocol is introduced as a novel MAC protocol combining CSMA and S-TDMA for multimedia communication with optimal utilization for a motorway environment. This proposed protocol was integrated in the simulator to allow managing the access to the base station. In 'Behavior of Clock-Sampling Mutual Network Synchronization in Wireless Sensor Networks: Convergence and Security', MacNeil et al. presented some simulation results that indicate the potential of clock sampling mutual network synchronization algorithm (CS-MNS) to achieve high clock synchronization accuracy in mobile multi-hop wireless networks. It is shown through analysis that in the absence of offset errors the network clocks converge. Finally, a method for adding external reference synchronization that is compatible with a security discussion, given in the paper, is proposed. Finally, we would like to express our gratitude to the Editor-in-Chief, Dr. Mohsen Guizani for his advice, patience, and encouragements since the beginning until the final stage. We thank all anonymous reviewers who spent much of their precious time reviewing all the papers. Their timely reviews and comments greatly helped us select the best papers in this special issue. We also thank all authors who have submitted their papers for consideration for this issue. A special thank goes to Jennifer Chichester Hillier, who made a great effort on the production of this issue within a very tight schedule. We hope you will enjoy reading the great selection of papers in this issue.

Wikis are currently used in business to provide knowledge management systems, especially for individual organizations. However, building wikis manually is a laborious and time-consuming work. To assist founding wikis, we propose a methodology in this paper to automatically select the best snippets for entities as their initial explanations. Our method consists of two steps. First, we focus on extracting snippets from a given set of web pages for each entity. Starting from a seed sentence, a snippet grows up by adding the most relevant neighboring sentences into itself. The sentences are chosen by the Snippet Growth Model, which employs a distance function and an influence function to make decisions. Secondly, we pick out the best snippet for each aspect of an entity. The combination of all the selected snippets serves as the primary description of the entity. We present three ever-increasing methods to handle selection process. Experimental results based on a real data set show that our proposed method works effectively in producing primary descriptions for entities such as employee names.

Previous anti-spamming algorithms based on link structure suffer from either the weakness of the page value metric or the vagueness of the seed selection. In this paper, we propose two page value metrics, AVRank and HVRank. These two "values" of all the web pages can be well assessed by using the bidirectional links' information. Moreover, with the help of bidirectional links, it becomes easier to enlarge the propagation coverage of seed sets. We further discuss the effectiveness of the combination of these two metrics, such as the quadratic mean of them. Our experimental results show that with such two metrics, our method can filter out spam sites and identify reputable ones more effectively than previous algorithms such as TrustRank.

Link analysis has been widely used to evaluate the importance of Web pages. Popular link analysis algorithms are mainly based on the link structure between pages. However, a Web page usually contains various links such as for navigation, decoration or nepotism, which are irrelevant to the topic of the Web page and can not reflect the actual voting relations between pages. In order to improve the performance of Web ranking, we bring out one filtering algorithm to recognize and eliminate these unrelated links using Content Lexical and Positional analysis. Experimental results on different Web domains show that our filtering model can efficiently detect the irrelevant links and effectively help to build a good link graph for the ranking calculation.

Seed selection is of significant importance for the biased PageRank algorithms such as TrustRank to combat link spamming. Previous work usually uses a small seed set, which has a big problem that the top ranking results have a strong bias towards seeds. In this paper, we analyze the relationship between the result bias and the number of seeds. Furthermore, we experimentally show that an automatically selected large seed set can work better than a carefully selected small seed set.

To bridge the increasing processor-disk performance gap, buffer caches are used in both storage clients (e.g. database systems) and storage servers to reduce the number of slow disk accesses. These buffer caches need to be managed effectively to deliver the performance commensurate to the aggregate buffer cache size. To address this problem, two paradigms have been proposed recently to collaboratively manage these buffer caches together: the hierarchy-aware caching maintains the same I/O interface and is fully transparent to the storage client software, and the aggressively-collaborative caching trades off transparency for performance and requires changes to both the interface and the storage client software. Before storage industry starts to implement collaborative caching in real systems, it is crucial to find out whether sacrificing transparency is really worthwhile, i.e., how much can we gain by using