Open-Set Domain Generalization (OSDG) aims to generalize over unseen target domains containing open classes, and the core challenge lies in identifying unknown samples never encountered during training. Recently, CLIP has exhibited impressive performance in OSDG, while it still falls into the dilemma between structural risk of known classes and open space risk from unknown classes, and easily suffers from over-confidence, especially when distinguishing known-like unknown samples. To this end, we propose a Semantic-enhanced CLIP (SeeCLIP) framework that leverages fine-grained semantics to boost unknown detection, so as to accommodate both risks and enable precise discrimination among categories. In SeeCLIP, we propose a semantic-aware prompt enhancement module to extract fine-grained key semantic features, and establish a fine-grained vision-language alignment. Duplex contrastive learning is proposed for prompt learning, which jointly optimizes duplex losses such that the unknown prompt is similar to known prompts, yet exhibits key semantic differences. We also design a semantic-guided diffusion module to enable nuanced capture in generation. By injecting perturbed key semantics into a diffusion model as control conditions, it generates the closest unknowns or pseudo-open samples with high similarity yet low belongingness to known classes. We formulate a generalization bound for OSDG, and show that SeeCLIP can achieve a lower generalization risk. Extensive experiments on benchmark datasets validate the superiority of SeeCLIP, it outperforms the SOTA methods by nearly 3% on accuracy and 5% on H-index, respectively.

Graph similarity computation (GSC) is a fundamental research problem in the field of graph learning evolving from early A*-based approximation algorithms to more efficient methods based on graph neural networks. The latter predicts similarity scores presented in the form of graph edit distance or maximum common subgraph by generating embeddings of interaction information between graph pairs. However, current methods often struggle to represent and match the inherent sequential structures in graphs, e.g., paths, which are crucial for the performance of GSC task. To address this issue, we propose a novel learning-based GSC method that incorporates path sampling, encoding, and matching in an end-to-end framework. By selecting critical paths via a centrality-based sampling strategy and learning path-level representations through cross-graph interactions, our method enables effective sequential structure matching between graphs. Extensive experiments on benchmark datasets show that our method achieves lower mean squared error on structure-sensitive datasets compared to existing leading methods and also achieves state-of-the-art performance on key ranking metrics including Spearman’s$\rho$, Kendall’s$\tau$, and Precision@10. Ablation studies further confirm the effectiveness of each proposed component.

Spectral Graph Neural Networks process graph signals using the spectral properties of the normalized graph Laplacian matrix. However, the frequent occurrence of repeated eigenvalues limits the expressiveness of spectral GNNs. To address this, we propose a higher-dimensional sheaf Laplacian matrix, which not only encodes the graph's topological information but also increases the upper bound on the number of distinct eigenvalues. The sheaf Laplacian matrix is derived from carefully designed perturbations of the block form of the normalized graph Laplacian, yielding a perturbed sheaf Laplacian (PSL) matrix with more distinct eigenvalues. We provide a theoretical analysis of the expressiveness of spectral GNNs equipped with the PSL and establish perturbation bounds for the eigenvalues. Extensive experiments on benchmark datasets for node classification demonstrate that incorporating the perturbed sheaf Laplacian enhances the performance of spectral GNNs.

Depression has currently become a serious social problem worldwide. However, the need for experienced doctors and tedious medical examinations greatly increases the inconvenience in diagnosing the depression. A convenient depression diagnosis method can significantly improve the medical experience of depression patients, and can greatly reduce the workload of doctors. In this paper, a Convenient Depression Diagnosis method based on Audio-Visual Deep Learning (CDD-AVDL) is proposed. CDD-AVDL exploits the videos of testers reading a specially-designed text, and note that the videos contain many subconscious human reactions (e.g., micro expressions, voice variations), which are difficultly affected by the artificial interventions, thus enabling the depression diagnosis results more accurate. In CDD-AVDL, the source features are first extracted from audios and visuals, and then the time-sequential features are extracted. Finally, a full connection layer and a convolution layer fusion are responsible for fusing the audio-visual features to yield the depression probabilities. Extensive experiments and clinical tests show that CDD-AVDL outperforms the state-of-the-arts in terms of the accuracy of depression diagnosis. Moreover, the data collection manner in CDD-AVDL is convenient, and the training cost of CDD-AVDL is very low.

Graph machine learning (GML) has made great progress in node classification, link prediction, graph classification, and so on. However, graphs in reality are often structurally imbalanced, that is, only a few hub nodes have a denser local structure and higher influence. The imbalance may compromise the robustness of existing GML models, especially in learning tail nodes. This article proposes a selective graph augmentation method to solve this problem. Firstly, a Pagerank-based sampling strategy is designed to identify hub nodes and tail nodes in the graph. Secondly, a selective augmentation strategy is proposed, which drops the noise neighbors of hub nodes on one side, and discovers the latent neighbors and generates pseudo neighbors for tail nodes on the other side. Also, it can alleviate the structural imbalance between two types of nodes. Finally, a GNN model is retrained on the augmented graph. Extensive experiments demonstrate that the proposed method can significantly improve the backbone GNNs and achieve superior performance to its competitors of graph augmentation methods and hub/tail aware methods.

Self-supervised learning on graphs has recently drawn a lot of attention due to its independence from labels and its robustness in representation. Current studies on this topic mainly use static information such as graph structures but cannot well capture dynamic information such as timestamps of edges. Realistic graphs are often dynamic, which means the interaction between nodes occurs at a specific time. This article proposes a self-supervised dynamic graph representation learning framework DySubC, which defines a temporal subgraph contrastive learning task to simultaneously learn the structural and evolutional features of a dynamic graph. Specifically, a novel temporal subgraph sampling strategy is firstly proposed, which takes each node of the dynamic graph as the central node and uses both neighborhood structures and edge timestamps to sample the corresponding temporal subgraph. The subgraph representation function is then designed according to the influence of neighborhood nodes on the central node after encoding the nodes in each subgraph. Finally, the structural and temporal contrastive loss are defined to maximize the mutual information between node representation and temporal subgraph representation. Experiments on five real-world datasets demonstrate that (1) DySubC performs better than the related baselines including two graph contrastive learning models and five dynamic graph representation learning models, especially in the link prediction task, and (2) the use of temporal information cannot only sample more effective subgraphs, but also learn better representation by temporal contrastive loss.

With the rapid growth of scientific papers, understanding the changes and trends in a research area is rather time-consuming. The first challenge is to find related and comparable articles for the research. Comparative citations compare co-cited papers in a citation sentence and can serve as good guidance for researchers to track a research area. We thus go through comparative citations to find comparable objects and build a comparative scientific summarization corpus (CSSC). And then, we propose the comparative graph-based summarization (CGSUM) method to create comparative summaries using citations as guidance. The comparative graph is constructed using sentences as nodes and three different relationships of sentences as edges. The relationship that sentences occur in the same paper is used to calculate the salience of sentences, the relationship that sentences occur in two different papers is used to calculate the difference between sentences, and the relationship that sentences are related to citations is used to calculate the commonality of sentences. Experiments show that CGSUM outperforms comparative baselines on CSSC and performs well on DUC2006 and DUC2007.

Movie investors seek for high box-office revenue. Usually, it is not an easy task for investors to estimate the return on their invests for movies, due to the complicated factors that could impact the box-office revenue, such as movie stars' appeal, potential audience reactions, movie genre, and so on. In this paper, we design and implement SMAS, an investor-oriented Social Media Analysis System focusing on movie invests, which provides various modules for capturing public opinions, assessing the value of movie stars, analyzing the temporal changes of their box-office impact, and predicting box-office revenues.

This paper studies the problem of relationship prediction in heterogeneous information networks. Our goal is not only to predict links/relationships more accurately but also to provide more viable paths to facilitate the formation of new links/relationships. A relationship prediction method based on multi-label learning named ML3P is proposed. In ML3P, each meta-path between nodes is regarded as a type of relationship and is given a label. Under the framework of multi-label learning, any potential relationship including the target relationship can be predicted. The results of comparative experiments in DBLP and Twitter datasets show that ML3P better uses heterogeneous information in supervised learning process and thus achieves better performance. Moreover, our method can output the correlation between relationships.

A Personal Assistant (PA) agent is a software agent capable of helping people to handle tasks in their workplace. The paper proposes a memory mechanism for personal assistant agents in order to enhance the agent intelligence while working with the user or other agents. A brief state of the art concerning personal assistant agents is presented first and the necessity of having a memory mechanism is explained later. In contrast to some previous work, this paper attempts to construct a memory model based on a model of episodic memory. Furthermore, a memory mechanism is designed for personal assistant agents in a multi-agent platform. Finally, future work is considered to achieve this objective in a research and development (R&D) context.

Relevance feedback is an effective scheme bridging the gap between high-level semantics and low-level features in content-based image retrieval (CBIR). In contrast to previous methods which rely on labeled images provided by the user, this article attempts to enhance the performance of relevance feedback by exploiting unlabeled images existing in the database. Concretely, this article integrates the merits of semisupervised learning and active learning into the relevance feedback process. In detail, in each round of relevance feedback two simple learners are trained from the labeled data, that is, images from user query and user feedback. Each learner then labels some unlabeled images in the database for the other learner. After retraining with the additional labeled data, the learners reclassify the images in the database and then their classifications are merged. Images judged to be positive with high confidence are returned as the retrieval result, while those judged with low confidence are put into thepoolwhich is used in the next round of relevance feedback. Experiments show that using semisupervised learning and active learning simultaneously in CBIR is beneficial, and the proposed method achieves better performance than some existing methods.

In multi-instance learning, the training examples are bags composed of instances without labels and the task is to predict the labels of unseen bags through analyzing the training bags with known labels. In content-based image retrieval (CBIR), the query is ambiguous because it is hard to ask the user precisely specify what he or she wants. Such kind of ambiguity can be gracefully dealt with by multi-instance learning techniques, and previous research shows that bag generators can significantly influence the performance of a CBIR system. In this paper, a novel bag generator named ImaBag is presented, where the pixels of each image are first clustered based on their color and spatial features and then the clustered blocks are merged and converted into a specific number of instances. Experiments show that ImaBag achieves comparable results to some existing bag generators but is more efficient in retrieving images from databases.