Accurate multivariate time series forecasting (MTSF) is critical for intelligent web services in Web of Things. When confronted with unseen multivariate time series (MTS), the industry typically invests significant time and resources in training multiple models to identify the optimal model for deployment. This paper proposes a novel, efficient, and scalable MTSF model selection method that directly selects suitable MTSF methods based on data characteristics without extensive model training. Model selection is a core component of AutoML, which has made significant progress in recent years. However, existing methods incur high operational costs and cannot be directly applied to MTSF tasks. Moreover, there is a lack of a comprehensive and cohesive public time series library for MTSF model selection. To address these challenges, we compile the first large heterogeneous labeled MTSF model selection dataset, called the ModelPile, which covers 41 mainstream datasets across 11 domains. We then propose AutoMTSF, a large model-enabled model selection method that transforms the MTSF model selection problem into a time series classification problem and utilizes the ModelPile to unlock large-scale multi-dataset training. AutoMTSF first uses the pre-trained large model to encode raw MTS. Given the coarse-grained limitations of large model encoding, Recursive Temporal Pattern Feature (RTPF) is proposed to capture both fine-grained and global temporal feature evolution, thereby effectively mapping data characteristics to the MTSF method space. Experiments comparing AutoMTSF with 2 baselines, 17 MTSF methods, and 4 large time series models show that AutoMTSF outperforms state-of-the-art methods while maintaining comparable execution time. This work represents a critical step in validating the accuracy and efficiency of large model-enabled classification for MTSF.

Subgraph Anomaly Detection (SAD) is crucial for identifying groups that deviate from the regular pattern within graphs, which benefits different domains such as financial fraud and network security. However, current studies rely on traditional node detection methods and fixed sampling strategies of subgraph structures, which makes it difficult to learn the pattern collaboration behavior of subgraphs. To address this limitation, this paper proposes a novel unsupervised framework named PC-SAD. The PC-SAD framework first employs an improved Graph AutoEncoder to identify core anomaly nodes by capturing multi-scale neighborhood information. Starting from these core anomaly nodes, we sample candidate subgraphs with path, tree, and cyclic structures, and enhance them according to the characteristics of the subgraph structures. Subsequently, candidate subgraphs are fed into the proposed Pattern Collaboration-based Graph Contrastive Learning method to generate collaborative pattern embeddings, thereby distinguishing anomaly subgraphs. The experimental results show that PC-SAD outperforms the state-of-the-art baseline methods on four benchmark datasets, which proves that PC-SAD is an effective solution to detect anomaly subgraphs.

Graph Neural Networks (GNNs) are inherently vulnerable to link stealing attacks, as their structural aggregation mechanisms may inadvertently leak training graph data. Existing link stealing methods primarily rely on posterior similarity for inference but suffer from critical limitations: inherent semantic bias (e.g., misclassifying semantically similar but unconnected nodes) and insufficient structural information, which constrain attack performance. To address these issues, we propose TLSA (Transfer Learning-based Link Stealing Attack), a novel framework that captures generalized structural knowledge from multi-domain heterogeneous graphs based on cross-domain knowledge transfer, and merges it with posterior similarity to enhance attack performance. The cross-domain knowledge transfer is enabled by integrating partially leaked target subgraphs with shadow graphs. Specifically, TLSA designs a triple-level alignment mechanism, including node feature reconstruction, which unifies heterogeneous posterior dimensions across domains; trainable hub nodes with gradient-driven topological optimization, forming bidirectional learning loops that bridge target and shadow domains; and domain adversarial training, which minimizes graph distribution distance and ensures deep semantic consistency across domains. Since its extracted structure-aware features are fused with node-pair semantic similarity, TLSA generates enhanced attack features for accurate edge existence prediction, significantly improving link stealing performance. Extensive experiments on diverse graph datasets validate the effectiveness of TLSA.

Event Causality Identification (ECI) aims to detect whether there exists a causal relation between two events in a document.Existing studies adopt a kind of identifying after learning paradigm, where events' representations are first learned and then used for the identification.Furthermore, they mainly focus on the causality existence, but ignore causal direction.In this paper, we take care of the causal direction and propose a new identifying while learning mode for the ECI task.We argue that a few causal relations can be easily identified with high confidence, and the directionality and structure of these identified causalities can be utilized to update events' representations for boosting next round of causality identification.To this end, this paper designs an iterative learning and identifying framework: In each iteration, we construct an event causality graph, on which events' causal structure representations are updated for boosting causal identification.Experiments on two public datasets show that our approach outperforms the state-of-theart algorithms in both evaluations for causality existence identification and direction identification. 1

Event Causality Identification (ECI) aims at determining the existence of a causal relation between two events.Although recent prompt learning-based approaches have shown promising improvements on the ECI task, their performance are often subject to the delicate design of multiple prompts and the positive correlations between the main task and derivate tasks.The in-context learning paradigm provides explicit guidance for label prediction in the prompt learning paradigm, alleviating its reliance on complex prompts and derivative tasks.However, it does not distinguish between positive and negative demonstrations for analogy learning.Motivated from such considerations, this paper proposes an In-Context Contrastive Learning (ICCL) model that utilizes contrastive learning to enhance the effectiveness of both positive and negative demonstrations.Additionally, we apply contrastive learning to event pairs to better facilitate event causality identification.Our ICCL is evaluated on the widely used corpora, including the EventStoryLine and Causal-TimeBank, and results show significant performance improvements over the state-of-the-art algorithms. 1

Ideology detection (ID) is important for gaining insights about peoples' opinions and stances on our world and society, which can find many applications in politics, economics and social sciences.It is not uncommon that a piece of text can contain descriptions of various issues.It is also widely accepted that a person can take different ideological stances in different facets.However, existing datasets for the ID task only label a text as ideologically left-or right-leaning as a whole, regardless whether the text containing one or more different issues.Moreover, most prior work annotates texts from data resources with known ideological bias through distant supervision approaches, which may result in many false labels.With some theoretical help from social sciences, this work first designs an ideological schema containing five domains and twelve facets for a new multifaceted ideology detection (MID) task to provide a more complete and delicate description of ideology.We construct a MITweet dataset for the MID task, which contains 12,594 English Twitter posts, each annotated with a Relevance and an Ideology label for all twelve facets.We also design and test a few of strong baselines for the MID task under in-topic and cross-topic settings, which can serve as benchmarks for further research.

Implicit Discourse Relation Recognition (IDRR) is to detect and classify relation sense between two text segments without an explicit connective. Vanilla pre-train and fine-tuning paradigm builds upon a Pre-trained Language Model (PLM) with a task-specific neural network. However, the task objective functions are often not in accordance with that of the PLM. Furthermore, this paradigm cannot well exploit some linguistic evidence embedded in the pre-training process. The recent pre-train, prompt, and predict paradigm selects appropriate prompts to reformulate downstream tasks, so as to utilizing the PLM itself for prediction. However, for its success applications, prompts, verbalizer as well as model training should still be carefully designed for different tasks. As the first trial of using this new paradigm for IDRR, this paper develops a Connective-cloze Prompt (ConnPrompt) to transform the relation prediction task as a connective-cloze task. Specifically, we design two styles of ConnPrompt template: Insert-cloze Prompt (ICP) and Prefix-cloze Prompt (PCP) and construct an answer space mapping to the relation senses based on the hierarchy sense tags and implicit connectives. Furthermore, we use a multi-prompt ensemble to fuse predictions from different prompting results. Experiments on the PDTB corpus show that our method significantly outperforms the state-of-the-art algorithms, even with fewer training data.

Recently, prompt-tuning has attracted growing interests in event argument extraction (EAE).However, the existing prompt-tuning methods have not achieved satisfactory performance due to the lack of consideration of entity information.In this paper, we propose a bidirectional iterative prompt-tuning method for EAE, where the EAE task is treated as a clozestyle task to take full advantage of entity information and pre-trained language models (PLMs).Furthermore, our method explores event argument interactions by introducing the argument roles of contextual entities into prompt construction.Since template and verbalizer are two crucial components in a clozestyle prompt, we propose to utilize the role label semantic knowledge to construct a semantic verbalizer and design three kinds of templates for the EAE task.Experiments on the ACE 2005 English dataset with standard and low-resource settings show that the proposed method significantly outperforms the peer stateof-the-art methods.Our code is available at https://github.com/HustMinsLab/BIP.

Social emotion classification is to estimate the distribution of readers' emotion evoked by an article. In this paper, we design a new neural network model by encoding sentence syntactic dependency and document topical information into the document representation. We first use a dependency embedded recursive neural network to learn syntactic features for each sentence, and then use a gated recurrent unit to transform the sentences' vectors into a document vector. We also use a multi-layer perceptron to encode the topical information of a document into a topic vector. Finally, a gate layer is used to compose the document representation from the gated summation of the document vector and the topic vector. Experiment results on two public datasets indicate that our proposed model outperforms the state-of-the-art methods in terms of better average Pearson correlation coefficient and MicroF1 performance.