Domain generalizable person Re-identification methods (DG person ReID) often focus on aligning source-domain distributions to learn domain-invariant features. However, they commonly overlook the pivotal role of hard samples and subtle yet important appearance differences across domains in refining decision boundaries, leading to suboptimal performance when encountering visually similar pedestrians. To address these issues, we propose a framework that combines adaptive uncertainty-driven hard identity mining with a multi-level feature propagation strategy. Specifically, we first identify and prioritize challenging samples across domains through an uncertainty estimation mechanism coupled with dynamic cross-domain weight allocation, thereby adaptively enhancing the model’s discriminative capability for confusable identities. We then incorporate both cross-domain and intra-domain feature propagation to integrate fine-grained information across multiple source domains, strengthening the model’s adaptability to unseen target domains. Extensive experiments on nine real-world benchmarks demonstrate that our method consistently outperforms state-of-the-art DG person ReID approaches, achieving up to 5.3% mAP and 6.6% Rank-1 improvements over the leading baselines.

Single Domain Generalization (SDG) aims to develop models capable of generalizing to unseen target domains using only one source domain, a task complicated by substantial domain shifts and limited data diversity. Existing SDG approaches primarily rely on data augmentation techniques, which struggle to effectively adapt training dynamics to accommodate large domain shifts. To address this, we propose LEAwareSGD, a novel Lyapunov Exponent (LE)-guided optimization approach inspired by dynamical systems theory. By leveraging LE measurements to modulate the learning rate, LEAwareSGD encourages model training near the edge of chaos, a critical state that optimally balances stability and adaptability. This dynamic adjustment allows the model to explore a wider parameter space and capture more generalizable features, ultimately enhancing the model's generalization capability. Extensive experiments on PACS, OfficeHome, and DomainNet demonstrate that LEAwareSGD yields substantial generalization gains, achieving up to 9.47\% improvement on PACS in low-data regimes. These results underscore the effectiveness of training near the edge of chaos for enhancing model generalization capability in SDG tasks.

Single domain generalization (SDG) aims to learn a generalizable model from only one source domain available to unseen target domains. Existing SDG techniques rely on data or feature augmentation to generate distributions that complement the source domain. However, these approaches fail to address the challenge where gradient conflicts from synthesized domains impede the learning of domain-invariant representation. Inspired by the concept of mechanical equilibrium in physics, we propose a novel conflict-aware approach named domain gradient equilibrium for SDG. Unlike prior conflict-aware SDG methods that alleviate the gradient conflicts by setting them to zero or random values, the proposed domain gradient equilibrium method first decouples gradients into domaininvariant and domain-specific components. The domain-specific gradients are then adjusted and reweighted to achieve equilibrium, steering the model optimization toward a domain-invariant direction to enhance generalization capability. We conduct comprehensive experiments on four image recognition benchmarks, and our method achieves an accuracy improvement of 2.94% in the PACS dataset over existing state-of-the-art approaches, demonstrating the effectiveness of our proposed approach.

Accurately segmenting polyps from colonoscopy images is essential for diagnosing colorectal cancer. Despite the tremendous success of the deep convolutional neural networks in automatic polyp segmentation, it suffers from domain shift issues, where the trained model yields performance deterioration on unseen test datasets. This paper proposes an illumination enhancement-based domain generalization approach to improve the generalization capability of the model on unseen test datasets and alleviate this issue. In particular, an image decomposition module (IDM) was developed to separate colonoscopy images into reflectance, local, and global illumination components. An illumination transform module (ITM) was proposed to augment images with different global illuminations by synthesizing target-like global illumination maps. A novel illumination variance insensitiveness (IViSen) is also introduced to evaluate the robustness of the model against illumination disturbance. IViSen is easy to compute and correlates well with model generalizability. The segmentation performance of the proposed model on four colonoscopy datasets was examined: CVC-ClinicDB, CVC-ColonDB, ETIS-Larib, and Kvasir-SEG. The method outperformed the competitive methods when tested on unseen domains. In particular, the proposed approach yielded 60.82% and 53.19% in terms of mean Dice and IoU, respectively, with 2.06% and 2.31% improvements.

Learning the next Point-of-Interest (POI) is a highly context-dependent human movement behavior prediction task, which has gained increasing attention with the consideration of massive spatial-temporal trajectories data or check-in data. The spatial dependency, temporal dependency, sequential dependency and social network dependency are widely considered pivotal to predict the users’ next location in the near future. However, most existing models fail to consider the influence of other users’ movement patterns and the correlation with the POIs the user has visited. Therefore, we propose a Global User Graph Enhanced Network (GUGEN) for the next POI recommendation from a global and a user perspectives. First, a trajectory learning network is designed to model the users’ short-term preference. Second, a geographical learning module is designed to model the global and user context information. From the global perspective, two graphs are designed to represent the global POI features and the geographical relationships of all POIs. From the user perspective, a user graph is constructed to describe each users’ historical POI information. We evaluated the proposed model on three real-world datasets. The experimental evaluations demonstrate that the proposed GUGEN method outperforms the state-of-the-art approaches for the next POI recommendation.

Crowd flow forecasting is of great significance for urban traffic management and personal travel planning. Due to the complexity of the urban geographic structure and the highly nonlinear temporal and spatial dependence on crowd flow, accurate forecasting becomes very challenging. Recent research works usually divided cities into regions of the same size and coded as heat-maps, for cities with complex terrain, heat-maps contain many invalid data, which have a negative effect on the acquisition of spatial dependence. In order to decrease the effect, we encode the crowd flow into graphs and propose a multi-graph neural network based model to solve the crowd flow forecasting problem. We first construct two K-NN graphs by the Euclidean distance and the Pearson correlation coefficient respectively and the spatial dependence is captured through the spatial block composed of Graph Attention Networks(GAT) and ChebNet, then another ChebNet is deployed to fuse the spatial dependency of the two graphs. Afterward, we adapted a LSTM to capture the temporal dependence of all regions separately and use self-attention mechanism and fully connected layer to (a) the receptive filed of CNN's filget prediction results. Extensive experiment results based on two real-world datasets demonstrate that our model achieves an important performance on other baselines.

Unlabeled data can be easily collected and help to exploit the correlations among different modalities. Existing works tried to explore label information contained in unlabeled data, however most of them suffer from difficulties in separating samples from different categories and have great interference. This paper proposes a novel method named semi-supervised cross-modal hashing based on label prediction and distance preserving(SS-LPDP). First, we use the deep neural networks to extract the feature of the labeled data among different modalities and get the feature distribution of each category. Second, the similarity of the data among different modalities is maximized based on the extracted feature and the label information. A common objective function is proposed with distance preserving constraint, which can effectively separate data into different categories and reduce interference in retrieval. An optimization algorithm is used to update the network parameters of feature learning in each modality, and the label information of unlabeled data are dynamically updated according to the changes of the feature distribution in each iteration. Experimental evaluation on Wiki, Pascal and NUS-WIDE datasets show that the proposed method outperforms recent methods when we set 25% samples without category labels.

Recently, Domain Adaptation has received great attention in addressing domain shift problem, where source domain and target domain follow different distribution. Existing methods often seek to minimize the marginal distribution and the conditional distribution across domains however treat them equally, which is not reasonable in real applications. Moreover, it is easy to misclassify on target data with large within-class scatter. In this paper, we propose a Compact Cluster-based Balanced Distribution Adaptation (CC-BDA) method for cross-domain classification. CC-BDA exploits the balanced domain adaptation by increasing the balance factor dynamically in each iteration until convergence. In addition, we construct compact clusters to reduce within-class scatter alongside the domain adaptation, which endows samples with the same label much closer from their means. Finally, we present analysis and performance evaluation over seven state-of-the-art baseline methods. The results show that CC-BDA significantly outperforms other adaptation methods in accuracy and execution time cost.

Recursive query processing has experienced a recent resurgence, as a result of its use in many modern application domains, including data integration, graph analytics, security, program analysis, networking and decision making. Due to the large volumes of data being processed, several research efforts across multiple communities have explored how to scale up recursive queries, typically expressed in Datalog. Our experience with these tools indicate that their performance does not translate across domains---e.g., a tool designed for large-scale graph analytics does not exhibit the same performance on program-analysis tasks, and vice versa. Starting from the above observation, we make the following two contributions. First, we perform a detailed experimental evaluation comparing a number of state-of-the-art Datalog systems on a wide spectrum of graph analytics and program-analysis tasks, and summarize the pros and cons of existing techniques. Second, we design and implement our own general-purpose Datalog engine, called RecStep, on top of a parallel single-node relational system. We outline the techniques we applied on RecStep, as well as the contribution of each technique to the overall performance. Using RecStep as a baseline, we demonstrate that it generally out-performs state-of-the-art parallel Datalog engines on complex and large-scale Datalog evaluation, by a 4-6X margin. An additional insight from our work is that it is possible to build a high-performance Datalog system on top of a relational engine, an idea that has been dismissed in past work.

Modern servers pack enough storage and computing power that just a decade ago was spread across a modest-sized cluster. This paper presents a prototype system, called Quickstep, to exploit the large amount of parallelism that is packed inside modern servers. Quickstep builds on a vast body of previous methods for organizing data, optimizing, scheduling and executing queries, and brings them together in a single system. Quickstep also includes new query processing methods that go beyond previous approaches. To keep the project focused, the project's initial target is read-mostly in-memory data warehousing workloads in single-node settings. In this paper, we describe the design and implementation of Quickstep for this target application space. We also present experimental results comparing the performance of Quickstep to a number of other systems, demonstrating that Quickstep is often faster than many other contemporary systems, and in some cases faster by orders-of-magnitude. Quickstep is an Apache (incubating) project.