While Large Language Models (LLMs) enable complex autonomous behavior, current agents remain constrained by static, humandesigned prompts that limit adaptability.Existing self-improving frameworks attempt to bridge this gap but typically rely on inefficient, multi-turn recursive loops that incur high computational costs.To address this, we propose Metacognitive Agent Reflective Self-improvement (MARS), a framework that achieves efficient self-evolution within a single recurrence cycle.Inspired by educational psychology, MARS mimics human learning by integrating principle-based reflection (abstracting normative rules to avoid errors) and procedural reflection (deriving step-by-step strategies for success).By synthesizing these insights into optimized instructions, MARS allows agents to systematically refine their reasoning logic without continuous online feedback.Extensive experiments on six benchmarks demonstrate that MARS outperforms state-of-the-art self-evolving systems while significantly reducing computational overhead.

Time-Series (TS) data has grown in importance with the rise of Internet of Things devices like sensors, but its labeling remains costly and complex. While Unsupervised Domain Adaptation (UDAs) offers an effective solution, growing data privacy concerns have led to the development of Source-Free UDA (SFUDAs), enabling model adaptation to target domains without accessing source data. Despite their potential, applying existing SFUDAs to TS data is challenging due to the difficulty of transferring temporal dependencies-an essential characteristic of TS data-particularly in the absence of source samples. Although prior works attempt to address this by specific source pretraining designs, such requirements are often impractical, as source data owners cannot be expected to adhere to particular pretraining schemes. To address this, we propose Temporal Source Recovery (TemSR), a framework that leverages the intrinsic properties of TS data to generate a source-like domain and recover source temporal dependencies. With this domain, TemSR enables dependency transfer to the target domain without accessing source data or relying on source-specific designs, thereby facilitating effective and practical TS-SFUDA. TemSR features a masking-recovery-optimization process to generate a source-like distribution with restored temporal dependencies. This distribution is further refined through local context-aware regularization to preserve local dependencies, and anchor-based recovery diversity maximization to promote distributional diversity. Together, these components enable effective temporal dependency recovery and facilitate transfer across domains using standard UDA techniques. Extensive experiments across seven TS tasks demonstrate the effectiveness of TemSR, which even surpasses existing TS-SFUDA methods that require source-specific designs.

Functional brain connectivity networks capture complex relationships and temporal evolution between brain regions, which have become increasingly important for diagnosing neurological disorders. However, existing methods, which are primarily based on vector or graph representations, struggle to adequately characterize the intricate spatio-temporal topological architecture of functional brain networks. Additionally, they predominantly rely on data-driven paradigms and lack priors pertaining to cross-windows network interactions. To address these issues, we propose a spatio-temporal hypergraph attention network framework for brain network analysis. Specifically, we first propose a temporal attention network architecture embedded with temporal similarity-driven prior knowledge, which effectively extracts long-range dependency information from fMRI by combining multi-head self-attention mechanisms and cross-window temporal prior knowledge. Second, we design a hierarchical hypergraph generation module that fuses local and global brain topological information to achieve multi-scale modeling of high-order spatio-temporal structures. Additionally, the spatial attention network, developed based on transformer architecture, employs hypergraph message passing mechanisms to effectively construct multi-level spatial interaction relationships between brain regions. Finally, a multi-layer perceptron (MLP) is adopted for classification. Experiments on the ADNI and PD datasets demonstrate that our method outperforms several state-of-the-art approaches in diagnostic performance and provides discriminative graph features for relevant brain disease diagnosis.

Source-free domain adaptation (SFDA) adapts a pre-trained model from a labeled source domain to an unlabeled target domain without source data access, preserving privacy. While SFDA is common in computer vision, it remains largely unexplored in time series analysis, where existing methods struggle to capture temporal dynamics and often produce overconfident predictions on out-of-distribution samples. We propose MAsk And imPUte (MAPU), which tackles temporal consistency through a novel imputation task, where randomly masked time series signals are recovered within the learned embedding space. During adaptation, a dedicated temporal imputer guides the target model to generate features that maintain temporal consistency with source features. However, MAPU relies on standard softmax predictions, leading to overconfident predictions on target samples that fall outside the source domain's support. To address this limitation, we introduce Evidential-MAPU (E-MAPU), which leverages evidential uncertainty estimation to identify these out-of-support samples and adapts the feature extractor to map them closer to the source domain's support, while maintaining the classifier fixed. Extensive experiments on five real-world time series datasets demonstrate significant performance improvements over existing methods. Our approaches effectively handle various time series domain adaptation challenges while maintaining computational efficiency, achieving state-of-the-art performance through its uncertainty-aware adaptation strategy.

Source-Free Domain Adaptation (SFDA) aims to adapt a pre-trained model from an annotated source domain to an unlabelled target domain without accessing the source data, thereby preserving data privacy. While existing SFDA methods have proven effective in reducing reliance on source data, they struggle to perform well on multivariate time series (MTS) due to their failure to consider the intrinsic spatial correlations inherent in MTS data. These spatial correlations are crucial for accurately representing MTS data and preserving invariant information across domains. To address this challenge, we propose Temporal Restoration and Spatial Rewiring (TERSE), a novel and concise SFDA method tailored for MTS data. Specifically, TERSE comprises a customized spatial-temporal feature encoder designed to capture the underlying spatial-temporal characteristics, coupled with both temporal restoration and spatial rewiring tasks to reinstate latent representations of the temporally masked time series and the spatially masked correlated structures. During the target adaptation phase, the target encoder is guided to produce spatially and temporally consistent features with the source domain by leveraging the source pre-trained temporal restoration and spatial rewiring networks. Therefore, TERSE can effectively model and transfer spatial-temporal dependencies across domains, facilitating implicit feature alignment. In addition, as the first approach to simultaneously consider spatial-temporal consistency in MTS-SFDA, TERSE can also be integrated as a versatile plug-and-play module into established SFDA methods. Extensive experiments on three real-world time series datasets demonstrate the effectiveness and versatility of our approach. Our code is available at https://github.com/Tokenmw/TERSE-master.

Test-time adaptation aims to adapt pre-trained deep neural networks using solely online unlabelled test data during inference. Although TTA has shown promise in visual applications, its potential in time series contexts remains largely unexplored. Existing TTA methods, originally designed for visual tasks, may not effectively handle the complex temporal dynamics of real-world time series data, resulting in suboptimal adaptation performance. To address this gap, we propose Augmented Contrastive Clustering with Uncertainty-aware Prototyping (ACCUP), a straightforward yet effective TTA method for time series data. Initially, our approach employs augmentation ensemble on the time series data to capture diverse temporal information and variations, incorporating uncertainty-aware prototypes to distill essential characteristics. Additionally, we introduce an entropy comparison scheme to selectively acquire more confident predictions, enhancing the reliability of pseudo labels. Furthermore, we utilize augmented contrastive clustering to enhance feature discriminability and mitigate error accumulation from noisy pseudo labels, promoting cohesive clustering within the same class while facilitating clear separation between different classes. Extensive experiments conducted on three real-world time series datasets demonstrate the effectiveness and generalization potential of the proposed method, advancing the underexplored realm of TTA for time series data. Our code is available at https://github.com/Tokenmw/ACCUP-main.

EEG-based emotion recognition is crucial for understanding human affective states, offering valuable insights into diverse fields like mental health monitoring and humancomputer interaction. Recent advancements in graph learning have significantly impacted EEG emotion recognition due to their ability to model the complex, dynamic relationships within brain networks. However, current methods often neglect the interplay between shared and individual correlations among EEG channels. Furthermore, individual variations in EEG patterns lead to distributional shifts that hinder the generalization of existing approaches. This paper proposes a novel Two-stage domain Adaptation with Hybrid Adaptive Graph learning (TAHAG) for EEG emotion recognition. TAHAG first employs hybrid adaptive graph learning to capture both shared and individual spatial characteristics of the EEG signals, dynamically integrating their contributions. Feature attention mechanisms are then incorporated to refine node features and enhance the model's discriminability. To address distributional variations, TAHAG utilizes a two-stage domain adaptation strategy. This strategy involves aligning the refined node features across different domains through discrepancy alignment. Subsequently, adversarial training captures domain-invariant summarized features of the entire graph. Extensive experiments on three public datasets demonstrate the superiority of TAHAG compared to existing methods. Furthermore, visualization of neuronal activity reveals significant brain regions and inter-channel relationships relevant to EEG emotion recognition.

Domain adaptation is crucial in aerial imagery, as the visual representation of these images can significantly vary based on factors such as geographic location, time, and weather conditions. Additionally, high-resolution aerial images often require substantial storage space and may not be readily accessible to the public. To address these challenges, we propose a novel Source-Free Object Detection (SFOD) method. Specifically, our approach begins with a self-training framework, which significantly enhances the performance of baseline methods. To alleviate the noisy labels in self-training, we utilize Contrastive Language-Image Pre-training (CLIP) to guide the generation of pseudo-labels, termed CLIP-guided Aggregation (CGA). By leveraging CLIP’s zero-shot classification capability, we aggregate its scores with the original predicted bounding boxes, enabling us to obtain refined scores for the pseudo-labels. To validate the effectiveness of our method, we constructed two new datasets from different domains based on the DIOR dataset, named DIOR-C and DIOR-Cloudy. Experimental results demonstrate that our method outperforms other comparative algorithms. The code is available at https://github.com/Lans1ng/SFOD-RS.

Fatigue driving is a significant cause of road traffic accidents and associated casualties. Automatic assessment of driver drowsiness by monitoring electroencephalography (EEG) signals offer a more objective way to improve driving safety. However, most existing measures are based on multi-channel EEG signals, which are more difficult to apply in practical scenarios as it usually lacks better portability and comfort. In addition, due to the relatively parsimonious and non-stationary characteristics, it is still challenging to effectively accomplish drowsiness recognition by exploiting single-channel EEG signals alone. To this end, we propose a novel temporal-frequential attentional convolutional neural network (TFAC-Net) to take full advantage of spectral-temporal features for single-channel EEG driver drowsiness recognition. Specifically, to capture the potentially valuable information contained in single-channel EEG, the continuous wavelet transform is first employed to generate a corresponding spectral-temporal representation. Then, the temporal-frequential attention mechanism is adopted to reveal critical time-frequency regions in terms of the driver’s mental state. Finally, an adaptive feature fusion module is considered to recalibrate and integrate the most relevant feature channels for final prediction. Extensive experimental results on a widely used public EEG driving dataset demonstrate that the TFAC-Net approach is superior to the state-of-the-art methods, and could discover some discriminative temporal-frequential regions. Moreover, this study also sheds light on the development of portable EEG devices and practical driver drowsiness recognition.

Emotion recognition based on electroencephalography (EEG) has attracted significant attention and achieved considerable advances in the fields of affective computing and human-computer interaction. However, most existing studies ignore the coupling and complementarity of complex spatiotemporal patterns in EEG signals. Moreover, how to exploit and fuse crucial discriminative aspects in high redundancy and low signal-to-noise ratio EEG signals remains a great challenge for emotion recognition. In this paper, we propose a novel attention-based spatial-temporal dual-stream fusion network, named ASTDF-Net, for EEG-based emotion recognition. Specifically, ASTDF-Net comprises three main stages: first, the collaborative embedding module is designed to learn a joint latent subspace to capture the coupling of complicated spatiotemporal information in EEG signals. Second, stacked parallel spatial and temporal attention streams are employed to extract the most essential discriminative features and filter out redundant task-irrelevant factors. Finally, the hybrid attention-based feature fusion module is proposed to integrate significant features discovered from the dual-stream structure to take full advantage of the complementarity of the diverse characteristics. Extensive experiments on two publicly available emotion recognition datasets indicate that our proposed approach consistently outperforms state-of-the-art methods.