Large-scale EEG foundation models have shown strong generalization across a range of downstream tasks, but their training remains resource-intensive due to the volume and variable quality of EEG data. In this work, we introduce EEG-DLite, a data distillation framework that enables more efficient pre-training by selectively removing noisy and redundant samples from large EEG datasets. EEG-DLite begins by encoding EEG segments into compact latent representations using a self-supervised autoencoder, allowing sample selection to be performed efficiently and with reduced sensitivity to noise. Based on these representations, EEG-DLite filters out outliers and minimizes redundancy, resulting in a smaller yet informative subset that retains the diversity essential for effective foundation model training. Through extensive experiments, we demonstrate that training on only 5 percent of a 2,500-hour dataset curated with EEG-DLite yields performance comparable to, and in some cases better than, training on the full dataset across multiple downstream tasks. To our knowledge, this is the first systematic study of pre-training data distillation in the context of EEG foundation models. EEG-DLite provides a scalable and practical path toward more effective and efficient physiological foundation modeling.

We introduce a new analytical framework, developed based on the spatial network calculus, for performance assessment of Low Earth Orbit (LEO) satellite networks. Specifically, we model the satellites' spatial positions as a strong ball-regulated point process on the sphere. Under this model, proximal points in space exhibit a locally repulsive property, reflecting the fact that intersatellite links are protected by a safety distance and would not be arbitrarily close. Subsequently, we derive analytical lower bounds on the conditional coverage probabilities under Nakagami-$m$ and Rayleigh fading, respectively. These expressions have a low computational complexity, enabling efficient numerical evaluations. We validate the effectiveness of our theoretical model by contrasting the coverage probability obtained from our analysis with that estimated from a Starlink constellation. The results show that our analysis provides a tight lower bound on the actual value and, surprisingly, matches the empirical simulations almost perfectly with a 1 dB shift. This demonstrates our framework as an appropriate theoretical model for LEO satellite networks.

We investigate the role of channel access schemes in enhancing the timeliness of status updates in sensor networks. Specifically, we model the large-scale sensor network as a Poisson cellular network and derive the network average age of information (AoI) under five different channel access schemes: slotted ALOHA, frame slotted ALOHA, random scheduling, round robin, and channel-aware. These schemes are categorized based on random vs. scheduled access and non-channel-aware vs. channel-aware. Our goal is to investigate when the additional overhead and complexity introduced by scheduling and channel state information (CSI) are beneficial, enabling better decisions in network design. Our findings reveal that the effectiveness of these schemes is influenced by the signal-to-interference ratio (SIR) decoding threshold, which often reflects the length of communication data. For short-packet communications, the performance differences among various channel access strategies are minimal, and the gains from scheduling are limited. Additionally, the inclusion of extra CSI does not yield performance improvements; in fact, some simple scheduling strategies, along with channelaware strategy that leverage CSI, may not outperform basic random access methods. Among the protocols we examined, the round robin scheme achieves the best performance. In contrast, scheduled access schemes exhibit a clear performance advantage in long-packet communications. Furthermore, the channel-aware scheme significantly enhances the network AoI performance, particularly in networks with higher transmitter competition.

Robotic manipulation tasks often rely on visual inputs from cameras to perceive the environment. However, previous approaches still suffer from performance degradation when the camera’s viewpoint changes during manipulation. In this paper, we propose ReViWo (Representation learning for View-invariant World model), leveraging multi-view data to learn robust representations for control under viewpoint disturbance. ReViWo utilizes an autoencoder framework to reconstruct target images by an architecture that combines view-invariant representation (VIR) and view-dependent representation. To train ReViWo, we collect multi-view data in simulators with known view labels, meanwhile, ReViWo is simutaneously trained on Open X-Embodiment datasets without view labels. The VIR is then used to train a world model on pre-collected manipulation data and a policy through interaction with the world model. We evaluate the effectiveness of ReViWo in various viewpoint disturbance scenarios, including control under novel camera positions and frequent camera shaking, using the Meta-world & PandaGym environments. Besides, we also conduct experiments on real world ALOHA robot. The results demonstrate that ReViWo maintains robust performance under viewpoint disturbance, while baseline methods suffer from significant performance degradation. Furthermore, we show that the VIR captures task-relevant state information and remains stable for observations from novel viewpoints, validating the efficacy of the ReViWo approach.

Grounded situation recognition (GSR) is a comprehensive structured scene understanding task that predicts the salient activity (verb), entities (nouns) involved in the activity with their roles, as well as the corresponding bounding-box groundings of the entities from the given image. Existing I.I.D.-based methods for GSR are limited in their ability to recognize novel verb-noun combinations. To address this problem, in this paper, we novelly consider GSR as a Non-I.I.D. task and focus on learning verb-invariant and role-specific representations for verb and noun predictions. Based on the causality, a novel Contrastive Invariant Risk Minimization (CIRM) model for GSR is proposed. In the proposed CIRM, invariant risk minimization is integrated into a transformer architecture to learn invariant representations for verb prediction. To enhance the intra-verb compactness and the inter-verb separability, contrastive learning is utilized to learn discriminative features. As far as we know, this is the first work that regards the task of GSR as a problem of out-of-distribution generalization. Extensive experiments on the benchmark SWiG dataset demonstrate the effectiveness of our proposed CIRM over other state-of-the-art methods in all evaluation metrics.

We investigate the role of channel access schemes in enhancing the timeliness of status updates in sensor networks. Specifically, we model the large-scale sensor network as a Poisson cellular network and derive the network average Age of Information (AoI) under two channel access schemes: random and scheduled access. Our findings reveal that the effectiveness of these schemes is influenced by the signal-to-interference ratio (SIR) decoding threshold, which often reflects the length of communication data. For short-packet communications, performance differences among various channel access strategies are minimal, and the gains from scheduling are limited; in fact, some simple scheduling strategies may not outperform basic random strategies. Conversely, scheduled access schemes demonstrate a distinct performance advantage for long-packet communications. The round robin scheme consistently yields the best performance among the four protocols we examined-slotted ALOHA, frame slotted ALOHA, random scheduling, and round robin scheduling. This is due to its ability to mitigate intra-cell interference and regularize both status updates and channel access periods for each sensor, which is particularly beneficial in reducing AoI.

Object: The effectiveness of simulated weightlessness on orthostatic endurance of healthy volunteers were observed by head-down-6° bed rest test.Method: The experiment was conducted in October 2020 at Space Science and Technology Institute (Shenzhen). A total of 9 volunteers were enrolled. Before -6° head-down bed rest and after getting up, HUT + LBNP (standing stress + lower body negative pressure) was performed on the volunteers. The time of standing stress and cardiac physiological function were observed, objective: to investigate the effect of simulated weightlessness on orthostatic endurance of volunteers.Result: After 15 days of -6° head-down bed-rest to simulate weightlessness, all volunteers' standing endurance time and cardiac physiological function indexes were worse than those before bed-rest during HUT + LBNP (P<0.05).Conclusion: -6° head-down bed rest for 15 days can reduce the volunteers standing endurance.

Object: To determine the efficacy of transcutaneous electrical acupoint stimulation (TEAS) at Neiguan (PC 6) in improving the stress response in standing-squatting test (SST).Method: According to the meridian theory of traditional Chinese medicine, the volunteers were given TEAS at PC6. To observe the changes of cardiac physiological function in volunteers during SST before and after stimulation, and to determine the effectiveness of improving stress response.Result: when the volunteers performed SST after the intervention of TEAS, HR was significantly lower than that before TEAS intervention( P<0.05 ); SBP, DBP and MBP were significantly increased compared with those before TEAS intervention. It is suggested that TEAS at PC 6 can improve the ability of cardiovascular regulation during SST stress stimulation.Conclusion: Transcutaneous electrical stimulation at PC 6 can improve the stress response ability of volunteers during standing-squatting test.

The development of wireless communication network technology has provided people with diversified and convenient services. However, with the expansion of network scale and the increase in the number of devices, malicious attacks on wireless communication are becoming increasingly prevalent, causing significant losses. Currently, wireless communication systems authenticate identities through certain data identifiers. However, this software-based data information can be forged or replicated. This article proposes the authentication of device identity using the hardware fingerprint of the terminal’s Radio Frequency (RF) components, which possesses properties of being genuine, unique, and stable, holding significant implications for wireless communication security. Through the collection and processing of raw data, extraction of various features including time-domain and frequency-domain features, and utilizing machine learning algorithms for training and constructing a legal fingerprint database, it is possible to achieve close to a 97% recognition accuracy for Fifth Generation (5G) terminals of the same model. This provides an additional and robust hardware-based security layer for 5G communication security, enhancing monitoring capability and reliability.

In recent years, open mobile social networks focused on socializing and dating purposes have gained widespread popularity, such as Soul, Tinder, Momo, and Tantan, among several others. These applications permit users to post, comment, and send private messages to other users without their consent, making communication accessible. However, this low-entry communication approach has also increased malicious user attacks. We delve into a comprehensive analysis of malicious accounts in open socializing and dating applications, revealing that the existing methods overlook hidden malicious signals within the user text-related information, thus resulting in poor detection performance. For such, we propose GraphTAM, a novel graph- and text-based multihead attention fusion network model for detecting such malicious accounts, consisting of modules that effectively combine nontext-related and text-related information, enhancing the accuracy and performance of detecting malicious accounts. We employ graph convolutional networks (GCNs) for nontext-related information to extract advanced representations of users, incorporating their attribute and social relationship features. Regarding text-related information, we employ a multihead attention model to identify suspicious patterns in users’ posted articles, comments, and relevant behavioral statistics, so finally, we merge the advanced representations of nontext-related and text-related information using a multilayer perceptron to determine the maliciousness of an account. Data sets collected from SLink are utilized for the experimental evaluation and to compare the performance of the proposed model with the several state of the art algorithms. Experimental results show significant advantages in malicious account detection, where the F1 score achieves over 0.9, outperforming the existing methods that range between 0.6 and 0.85. Furthermore, the comparative experiments substantiate the critical role of text-related information in detecting malicious accounts in open socializing and dating applications.

Pose-guided person image generation aims to synthesize person images in arbitrary poses. This task requires to perform spatial deformation on source images. Existing work often failed to transfer complex textures to generated images well. To solve this problem, we propose a novel network for this task. The network is called Cascaded Pyramid Deformation GAN(CPD-GAN), which can achieve more realistic results and conform more to the target person. In the extraction sub-network, the multi-scale feature modulation(MFM) blocks are proposed. A MFM block can fuse features at different scales into single scale feature. And in the genration sub-network, the dynamic transferring fusion(DTF) blocks are proposed to perform dynamic deformation on features from extraction sub-network and a cascaded pyramids structure is adopted to improve the quality of resulted complex textures. Experiments prove that our method performs better on pose transfer than other methods and utilizes each part of the network effectively.

Recently, very deep convolutional neural networks (CNNs) have shown great power in single image super-resolution (SISR) and achieved significant improvements against traditional methods. Among these CNN-based methods, the residual connections play a critical role in boosting the network performance. As the network depth grows, the residual features gradually focused on different aspects of the input image, which is very useful for reconstructing the spatial details. However, existing methods neglect to fully utilize the hierarchical features on the residual branches. To address this issue, we propose a novel residual feature aggregation (RFA) framework for more efficient feature extraction. The RFA framework groups several residual modules together and directly forwards the features on each local residual branch by adding skip connections. Therefore, the RFA framework is capable of aggregating these informative residual features to produce more representative features. To maximize the power of the RFA framework, we further propose an enhanced spatial attention (ESA) block to make the residual features to be more focused on critical spatial contents. The ESA block is designed to be lightweight and efficient. Our final RFANet is constructed by applying the proposed RFA framework with the ESA blocks. Comprehensive experiments demonstrate the necessity of our RFA framework and the superiority of our RFANet over state-of-the-art SISR methods.

Air-gapped networks achieve security by using the physical isolation to keep the computers and network from the Internet. However, magnetic covert channels based on CPU utilization have been proposed to help secret data to escape the Faraday-cage and the air-gap. Despite the success of such cover channels, they suffer from the high risk of being detected by the transmitter computer and the challenge of installing malware into such a computer. In this paper, we propose MagView, a distributed magnetic cover channel, where sensitive information is embedded in other data such as video and can be transmitted over the air-gapped internal network. When any computer uses the data such as playing the video, the sensitive information will leak through the magnetic covert channel. The "separation" of information embedding and leaking, combined with the fact that the covert channel can be created on any computer, overcomes these limitations. We demonstrate that CPU utilization for video decoding can be effectively controlled by changing the video frame type and reducing the quantization parameter without video quality degradation. We prototype MagView and achieve up to 8.9 bps throughput with BER as low as 0.0057. Experiments under different environment are conducted to show the robustness of MagView. Limitations and possible countermeasures are also discussed.