Unsupervised visible-infrared person reidentification (VI-ReID) is challenging due to the significant modality gap between visible and infrared images. Most existing methods rely on one-hot clustering pseudo-labels as supervision signals, which often fail to capture the full semantic relationships among samples and are highly susceptible to noise. To address these limitations, we propose a Semantic-aware Multimodal Collaborative Learning (SAMCL) framework for unsupervised VI-ReID. Specifically, a Modality-aware Semantic Fusion (MSF) module is designed to bridge the inter-modality gap by integrating complementary semantic details from both visible and infrared modalities, generating enriched cross-modal supervision signals, for cross-modal collaborative learning. Meanwhile, we present a Dynamic Contrastive Learning (DCL) module to refine intra-modality feature learning by dynamically aligning samples with their neighboring centroids in the feature space, improving clustering reliability and intra-modality feature discrimination. By combining the two modules, SAMCL harnesses multimodal collaboration, minimizes dependence on noisy pseudo-labels, and provides a robust approach to unsupervised VI-ReID. Extensive experiments demonstrate the superiority of our proposed method. For instance, on the SYSU-MM01 dataset, our model achieves a Rank-1 accuracy of 68.68% in the All Search setting, surpassing the state-of-the-art (SOTA) by 3.48%. On the RegDB dataset, it achieves a Rank-1 accuracy of 94.47% in the Visible-to-Infrared setting, outperforming the SOTA by 3.57%. On the LLCM dataset, it achieves a Rank-1 accuracy of 50.6% in the Visible-to-Infrared setting, outperforming the SOTA by 3.7%. The code is available at https://github.com/luoshixi123/SAMCL.

Event-based object detection plays a crucial role in scenarios involving high-speed motion, extreme lighting conditions, and high-frequency detection. However, existing methods fail to address the challenges posed by small objects, including discriminative feature deficiency, the loss of critical information, and the inherent sparsity of event data. Moreover, the lack of benchmark datasets has significantly hindered progress in this field. To tackle these issues, we propose the Fully Deformable Detection Network (FDDNet), a lightweight framework that dynamically adapts to extract key features. First, we introduce a Long-Term Deformable Temporal Receptive Module (LDTR), which aligns critical features across consecutive event streams and leverages a State Space Model for long-range temporal modeling, enhancing the detection of highspeed small objects. Second, to address the sparsity of event data and the concentration of key features along object edges, we design a Sparse Feature Aggregation Block (SFAB) within the backbone and a coarse-to-fine deformable detection head, enabling hierarchical feature refinement from local to global, and improving the detection quality of sparse targets. Finally, to mitigate the lack of event-based small object datasets, we develop a high-quality, annotation-free data acquisition method and collect a real-world benchmark dataset for validation. Extensive experiments demonstrate that our approach achieves state-of-the-art (SOTA) performance on event-based small object detection tasks, with a mAP of 37.4% (+2.4%) on our benchmark and runs at 88 FPS, showcasing both accuracy and real-time capability. Our code and Supplement are available at https://github.com/Lqm26/ESOD.

Multimodal sentiment analysis (MSA) provides a novel way to understand human sentiments. However, the differences between distribution patterns across modalities bring challenges in this domain. The inconsistency of recognitions with different modalities leads to incorrect final results. Moreover, the gaps between sentiments with different degrees are small in one modality, but the gaps between sentiments with same degree are large across different modalities. The imbalance leads to incorrect recognition for different sentiment degrees. Since the fuzzy network shows excellent performance in integrating data from multiple modalities, this study constructs a fuzzy cognition-based dynamic fusion network (Fcdnet) for MSA. The Fcdnet dynamically integrates sentiment scores across different modalities using a fuzzy cognition fusion mechanism, significantly enhancing the accuracy of identifying divergent sentiments across modalities. Additionally, a disparity balancing module is proposed to normalize the representations between different modality features by penalizing the similarity of sentiments with different degrees and rewarding the separability of sentiments with same degree. Experimental results demonstrate that Fcdnet outperforms state-of-the-art methods on public datasets, validating the superiority and effectiveness.

This study presents a systematic comparative analysis of Artificial Intelligence (AI)-enabled science education research by examining 59 peer-reviewed articles published between 2016 and 2025 from both domestic (Chinese) and international contexts. Through rigorous content analysis, we classify the literature according to publication trends, research themes, methodological approaches, educational settings, and principal findings. Overall publication volume has grown steadily, with domestic scholarship concentrating on theoretical constructs for personalized learning, AI-generated instructional resources, and Science, Technology, Engineering, and Mathematics (STEM) curriculum modelling, while international investigations emphasize empirical validation of automated assessment systems, generative AI applications, and large-scale efficacy evaluations. Methodologically, Chinese studies predominantly employ quantitative and framework-driven designs, whereas global research increasingly adopts mixed-methods and longitudinal approaches. Both strands focus primarily on primary and secondary (K-12) science education but diverge in subject emphasis and level of empirical rigour. Despite notable progress in intelligent tutoring systems, virtual inquiry environments, and data-driven personalization, domestic work trails behind in interdisciplinary integration, automated data collection, and longitudinal validation. We identify three key directions for future inquiry: (1) embedding multidisciplinary perspectives into AI-science education research, (2) automating data collection and measurement processes to enhance empirical robustness, and (3) expanding study contexts across educational stages and disciplines while establishing ethical AI frameworks to mitigate algorithmic bias and overreliance. These insights aim to inform the optimization of pedagogical practices, foster scientific literacy, and guide policy development in the digital era of education.

Large language models (LLMs) play a crucial role in intelligent code generation tasks. Most existing work focuses on pretraining or fine-tuning specialized code LLMs, e.g., CodeLlama. However, pretraining or fine-tuning a code LLM requires a vast corpus of data, significant computational resources, and considerable human effort. Compared to pretraining or fine-tuning LLMs, a simple and flexible method of contrastive decoding has garnered widespread attention to improve the text generation quality of LLMs. While contrastive decoding can indeed improve the text generation quality of LLMs, our research has found that directly using contrastive decoding: 1) introduces erroneous information into the logit distribution generated from normal prompts (i.e., user's input), particularly in the code generation of LLMs; 2) significantly impedes the inference and decoding time of LLMs. In this work, the limitations of using contrastive decoding directly are systematically highlighted, and a novel real-time fuzzy-assisted contrastive decoding (FCD) mechanism is proposed to improve the code generation quality of LLMs. The proposed FCD mechanism initially categorizes prompts into high-quality and low-quality groups based on the results of the evaluator (i.e., unit test) before integrating the LLM. Next, feature values (e.g., standard deviation, peak value, etc.) related to the logit distribution of predicted tokens during the LLM's inference process for both high-quality and low-quality prompts are extracted. Finally, the extracted feature values are used to train the fuzzy neural network (i.e, fuzzy min-max neural network) offline, allowing for the prejudgement of the reliability of the logit distribution for normal prompt outputs. This prevents the direct use of erroneous information from contrastive decoding and improves the code generation quality of LLMs. Through extensive experiments, it has been demonstrated that the proposed FCD mechanism can significantly improve the code generation quality of LLMs through FCD. Moreover, the FCD mechanism can also reduce the time required for inference and contrastive decoding.

Temporal action localization (TAL) is a significant and challenging task in the field of video understanding. It aims to locate the start and end timestamps of the actions in a video and recognize their categories. However, efficient action localization often requires extensive precise annotations. Therefore, the researchers propose weakly-supervised temporal action localization (WTAL), which aims to locate action instances in a video using only video level annotations. The existing WTAL methods lack the ability to distinguish the action context information effectively, including the pre-action and post-action scenes, which blur the action boundary and lead to the inaccurate action location. To solve the above problems, this paper proposes a global context-aware attention model (GCAM). Firstly, GCAM designs the mask attention module (MAM) to restrict the model's receptive field and make the model focus on localized features related to the action context. It enhances the ability to distinguish the action context information and clearly locate the start and end timestamps of the actions. Secondly, GCAM introduces the context broadcasting module (CBM), which supplements the global context information to keep the features intact in temporal dimension. This module solves the issue that the model overemphasizes the localized features due to the addition of the MAM. Extensive experiments on the THUMOS14 and ActivityNet1.2 datasets demonstrate the effectiveness of GCAM. On the THUMOS14 dataset, GCAM achieves an average mean average precision (mAP) of 49.5 %, representing a 2.2 % improvement over existing WTAL methods. On the ActivityNet1.2 dataset, GCAM achieves an average mAP of 27.2 %, representing a 0.3 % improvement over existing WTAL methods. These results highlight the superior performance of GCAM in accurately localizing actions in videos.

Deep learning (DL), as a cutting-edge technology in artificial intelligence, has significantly impacted fields such as computer vision and natural language processing. Loss function determines the convergence speed and accuracy of the DL model and has a crucial impact on algorithm quality and model performance. However, most of the existing studies focus on the improvement of specific problems of loss function, which lack a systematic summary and comparison, especially in computer vision and natural language processing tasks. Therefore, this paper reclassifies and summarizes the loss functions in DL and proposes a new category of metric loss. Furthermore, this paper conducts a fine-grained division of regression loss, classification loss, and metric loss, elaborating on the existing problems and improvements. Finally, the new trend of compound loss and generative loss is anticipated. The proposed paper provides a new perspective for loss function division and a systematic reference for researchers in the DL field.

In the modern sea battlefield, enemy corner reflectors can seriously affect the detection and identification capabilities of our radar systems. The corner reflector, through its unique geometry, greatly enhances the strength of the radar's echo, making it impossible for our radar to accurately and efficiently detect enemy ships. The high resolution range profile (HRRP) can distinguish the difference in distance between different scattering centers on the target, for which the ship target can be identified by the high resolution range profile of the ship and the corner reflector. However, due to the large difference in HRRP between the ship and the angular reflector at different angles, it is difficult to distinguish between different targets when the angle is sensitive to the change in target identification. Therefore, in this paper, we construct SAR images of ships and angular reflectors to recognize them in the distance and direction dimensions. In order to better deal with the problems of insufficient utilization of image structure information, high computational complexity and small samples brought about by the one-dimensional dimensionality reduction method, this paper proposes an improved recognition algorithm for ships and corner reflectors based on two-dimensional two dimensional two-way locally maintained projection, which uses 2DLDA+2DLPP to extract features in the column direction and 2DPCA+2DLPP to extract features in the row direction, and finally uses the nearest neighbor classifier for classification, which has a good recognition effect on a given dataset.

Computer vision is one of the most attractive research areas in artificial intelligence (AI), encompassing tasks such as object tracking, recognition, classification, and scene understanding [...]

Aim: DNA methylation in brain regions represents a potential mechanism linking genetic variation to Alzheimer’s disease (ad) risk, yet most studies have focused on blood-derived methylation markers. In this study, we conducted a systematic Mendelian randomization (MR) study to evaluate associations between predicted brain region-specific DNA methylation levels and ad risk, using methylation quantitative trait loci (mQTL) as genetic instruments. Methods: We analyzed mQTLs from five human brain regions: cerebellum (CRBLM), frontal cortex (FCTX), causal pons (PONS), and temporal cortex (TCTX) from 600 individuals in Gibbs et al’s study, as well as mQTLs from dorsolateral prefrontal cortex (DLPFC) of 543 participants in the Religious Orders Study and the Rush Memory and Aging Project (ROSMAP). In our MR analyses, we integrated these mQTLs with single nucleotide polymorphisms (SNP)-ad risk summary statistics derived from 85 934 ad-related cases and 401 577 normal controls. Results: Among 62 554 cytosine-guanine dinucleotide (CpG) sites, we identified 597 CpG sites (CpGs) significantly associated with ad risk (false discovery rate (FDR) < 0.05). Of these, 289 were confirmed through colocalization and summary-based MR (SMR) analyses, including one CpG site in CRBLM, 285 in DLPFC, one in FCTX, two in PONS, and one in TCTX. By integrating gene expression data, we identified 19 CpG sites with consistent associations across methylation levels, expression of eight target genes, and ad risk, including novel regulatory mechanisms involving RITA1’s modulation of cg11558705 and PCGF3’s regulation of cg10009224. Conclusion: Our findings highlight brain region-specific DNA methylation as a mediator of genetic risk for ad, offering insights into ad pathogenesis and identifying potential therapeutic targets.

U-Net has gained widespread application in image segmentation due to its symmetric structure design and skip connection mechanism. However, in the detection of small infrared targets, issues such as complex environmental backgrounds, small size of targets, and inherent characteristics (low signal-to-noise ratio, background clutter interference, low contrast between target and background) make it difficult for U-Net to effectively reconstruct target shapes during the decoding stage; this also exacerbates feature confusion, leading to an inability to achieve differentiated feature fusion. To address this issue, this paper proposes a non-symmetric context fusion attention U-Net. This method focuses on target responses and context during the encoding stage, extracting target features and contextual information at both shallow and deep levels. In the decoding stage, these extracted features are fused and gradually refined to enhance target response, yielding more accurate results. Experimental results show that CMAF demonstrates excellent performance on two public datasets.

Action detection and understanding provide the foundation for the generation and interaction of multimedia content. However, existing methods mainly focus on constructing complex relational inference networks, overlooking the judgment of detection effectiveness. Moreover, these methods frequently generate detection results with cognitive abnormalities. To solve the above problems, this study proposes a cognitive effectiveness network based on fuzzy inference (Cefdet), which introduces the concept of 'cognition--based detection' to simulate human cognition. First, a fuzzy-driven cognitive effectiveness evaluation module (FCM) is established to introduce fuzzy inference into action detection. FCM is combined with human action features to simulate the cognition-based detection process, which clearly locates the position of frames with cognitive abnormalities. Then, a fuzzy cognitive update strategy (FCS) is proposed based on the FCM, which utilizes fuzzy logic to re-detect the cognition-based detection results and effectively update the results with cognitive abnormalities. Experimental results demonstrate that Cefdet exhibits superior performance against several mainstream algorithms on the public datasets, validating its effectiveness and superiority.

In recent years, contrastive learning has made great progress in the field of computer vision, which shows great potential in action representation learning. Current contrastive learning methods usually employ contrastive loss function to learn video motion representation, which attend to capturing the similar background appearance but ignore similar motion information. This problem is called background bias, which restricts the model from exploring motion patterns. The background bias can be divided into wide background bias and micro background bias. The wide background bias refers to the statistically significant background bias, while the micro background bias refers to the background bias directly interacting with the moving object. To tackle these problems, this paper first proposes a semi-negative pair merging foreground–background, which mainly decouples the dynamic factor with obvious motion and the static factor with stable invariants in the video frame sequence. Then the dynamic factor of the original video is fused with other static factors to obtain a random background image where the foreground is more similar than the background to solve the wide background bias. Secondly, a pixel-level motion aware representation decomposition module calculates the pixel-level intensity variations from the feature space of adjacent frames, and these variations are further accumulated to obtain the salient map that focuses on the boundaries of moving objects, so the model attends to the motion pattern rather than the background to solve the micro background bias. Furthermore, a new dual loss function is proposed based on joint wide and micro background bias to better capture both static and dynamic features. The heat map shows that the proposed method effectively can solve background bias and achieve better performance in public datasets compared with other existing methods.

In the period of rapid development on the new information technologies, computer vision has become the most common application of artificial intelligence, which is represented by deep learning in the current society. As the most direct and effective application of computer vision, facial expression recognition (FER) has become a hot topic and used in many studies and domains. However, the existing FER methods focus on deep learning to generate increasingly complex attention structures, so they are unable to consider the connotative relationship between different parts of facial expressions. Moreover, the human expression recognition method based on complex deep learning network has serious interpretability issues. Therefore, in this paper, a novel Graph Neural Network (GNN) model is proposed to consider the systematic process of FER in human visual perception. Firstly, a region division mechanism is proposed, which divides the face region into six parts to unify the selection of key facial features. On this basis, in order to better consider the connotative relationship between different parts of facial expression, a human visual cognition strategy is proposed, which uses the divided six regions to learn facial expression features, and evenly selects the key features with high reliability as graph nodes. In combination with the human regional cooperative recognition process, the connotative relationship (such as relative position and similar structure) between graph nodes is extracted, so as to construct the GNN model. Finally, the effect of FER is obtained by the modeled GNN model. The experimental results compared with other related algorithms show that the model not only has stronger characterization and generalization ability, but also has better robustness compared with state-of-the-art methods.