Youliang Yuan, Qiuyang Mang, Jingbang Chen, Hong Wan, Xiaoyuan Liu, Junjielong Xu, Jen-tse Huang, Wenxuan Wang, Wenxiang Jiao, Pinjia He. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

mmWave radar-based human pose estimation is garnering increasing attention in Internet of Things (IoT) applications owing to its robustness under adverse lighting conditions and occlusion scenarios. However, accurately estimating human poses from mmWave signals remains challenging due to two key limitations: the lack of structural consistency when perceiving targets as a unified whole, and the absence of spatiotemporal consistency when tracking individual targets. To tackle these two challenges, this paper introduces RF-PoseR, a dedicated human pose rectifier for mmWave radar-based pose estimation. RF-PoseR serves as a corrective framework that refines the pose outputs generated by mmWave pose estimation systems. Based on the observation that perceptual inconsistencies in radar signals result in severely erroneous joint estimations, we introduce spatial and temporal constraints to enhance pose estimation accuracy. Specifically, RF-PoseR incorporates three core components: (1) unreliable joint removal guided by spatial structure, (2) pose completion leveraging temporal continuity, and (3) cross-modal pre-training utilizing large-scale visual pose datasets. Extensive experiments on mmWave radar pose datasets demonstrate that RF-PoseR significantly enhances the accuracy of poses generated by existing radar-based estimation networks. Furthermore, experiments on visual datasets confirm the method’s broader applicability beyond radar perception tasks.

Specific Emitter Identification (SEI) leverages unique hardware-induced Radio Frequency Fingerprints (RFFs) for secure physical-layer authentication. However, under cross-receiver scenarios where training and testing data exhibit hardware-induced distribution shifts, deep learning models are prone to shortcut learning. In such cases, networks inadvertently exploit spurious, receiver-specific artifacts as ”shortcuts” for identification rather than extracting the genuine, intrinsic fingerprints of the transmitter. To overcome this challenge, we propose a robust multi-task learning framework, termed MTL-SEI. This framework synergizes spectrum-based feature extraction with receiver-invariant adversarial training and channel-aware auxiliary supervision. Specifically, a gradient reversal layer (GRL) is employed to suppress receiver-dependent features, while an equalization-state prediction task provides semantic guidance to disentangle channel-induced distortions. Furthermore, an uncertainty-guided task weighting mechanism is introduced to dynamically balance the multiple optimization objectives based on predictive variance. Evaluations conducted on the ManySig dataset under a rigorous receiver-disjoint protocol demonstrate the superior generalization capability of MTL-SEI. Notably, our method achieves a transmitter identification accuracy of 88.50% —representing a 37.7% improvement over the 1D-CNN baseline—and yields an average performance gain of over 6.92% compared to state-of-the-art domain generalization methods. These results validate the effectiveness of the proposed feature disentanglement mechanism in mitigating receiver-induced biases.

Specific emitter identification (SEI) is a crucial technique for recognizing individual emitters based on the unique characteristics of their radio frequency signals. Deep learning (DL)-based SEI has become the dominant method for identifying and authenticating wireless devices. However, in real-world applications, electromagnetic signals are subject to dynamic channel conditions and variations across different receivers, leading to significant performance degradation when models trained on specific datasets are applied to new, unseen environments. This variability challenges traditional DL methods, making domain generalization (DG) an essential approach to tackle this issue. In this paper, we propose a robust SEI method, latent distribution mixing (LDM), to enhance model generalization in receiver-agnostic scenarios. Our approach first applies the Fourier transform to convert time-domain signals into frequency-domain representations. Then, it mixes latent feature distributions across domains in the feature space to improve robustness, enabling the model to adapt to domain shifts effectively. We evaluated our method on a cross-receiver dataset, achieving a peak performance of 88.52%, surpassing other domain generalization methods. The experimental results demonstrate that the proposed LDM method offers a promising solution for SEI tasks in cross-receiver scenarios. Our code can be downloaded from https://github.com/frownean/LDM.

As the global population continues to age and technologies such as the Internet of Things (IoT) and edge computing advance rapidly, indoor human activity recognition (HAR) based on Wi-Fi channel state information (CSI) has gained significant research attention. However, the high computational complexity of existing HAR methods limits their deployment on resource-constrained devices. To address this challenge, we propose a lightweight HAR method using branch decision lightweight two-stream convolution-augmented transformer (BLTHAT) model, which integrates depthwise separable convolutions (DSC) and an improved framework structure to enhance computational efficiency. Additionally, we introduce the branch fusion network (BFN), a decision-making module designed to optimize feature processing and improve model robustness. Further enhancements in attention mechanisms and regularization strategies contribute to reducing complexity while maintaining high recognition accuracy. Comprehensive experiments were conducted on a multi-label dataset. The results demonstrate that our proposed HAR method achieves high computational efficiency with minimal complexity, making it well-suited for IoT applications. Ablation studies further confirm that the multi-branch structure of the BFN module enhances feature extraction without significantly increasing computational overhead.

Radio Frequency Fingerprinting Identification (RFFI) leverages signal processing to extract unique characteristics from wireless signals for device identification. In recent years, deep learning (DL) has significantly advanced signal identification, catalyzing progress in RFFI research. This paper proposes an enhanced RFFI method to manage variable-length signal inputs, typically problematic for neural networks such as convolutional neural networks (CNNs) and multilayer perceptrons (MLPs), by treating these signals as images to solve data formatting problems. The robust representation of the variable-length signal ultimately achieves over 90% accuracy, meeting the expected results. Furthermore, conventional DL-based RFFI methods require a comprehensive analysis of the entire RF signal, consuming significant computational resources and vulnerable to environmental variations. We address these issues by proposing an incremental learning (IL)-based RFFI method that allows dynamic model updates and improves recognition and generalization performance. Our method’s efficacy, tested on the power amplifiers (PA) dataset, enables real-time data stream processing.

Specific emitter identification (SEI) is a technique that identifies individual emitters based on the inherent characteristics reflected in the radio frequency signals due to the individual differences of the emitters. Deep learning (DL) has become the primary research method for identifying and authenticating wireless devices in SEI. However, in the real world, electromagnetic signals continuously change with the channel environment and time, causing models trained on datasets collected from known specific domains to exhibit significant performance degradation when applied to unknown channel environments. This limitation makes general DL methods unsuitable, and domain generalization (DG) becomes a key method to address this issue. To overcome the limitations of SEI identification performance across different scenarios, we propose a robust SEI method by mixing augmentation, named SigMix. Specifically, we innovatively introduce the Mixup method into the SEI task, mixing data from different source domains and then performing pairwise linear interpolation before using it for training the neural network. The SigMix method helps the model learn more comprehensive features by generating new samples in the training data, thereby improving the model’s generalization ability. To validate the effectiveness of the SigMix method, while also considering the impact of different receivers on identification performance, we evaluate a dataset spanning both time and receivers. The experimental results indicate that the average identification accuracy of the proposed SigMix method in unknown domains reaches 84.40%, significantly outperforming existing DG methods, demonstrating the robustness and generalization of our proposed SigMix method in SEI tasks. Our code is available for download at://github.com/frownean/SigMix.

Drug-resistant epilepsy (DRE) patients typically require surgical intervention or neurostimulation. Therefore, accurate localization of the seizure onset zone (SOZ) is essential for effective clinical intervention. Although some physiologically meaningful parameters of neural computational models show substantial differences across brain regions during seizures, few studies pay attention to applying these model parameters to SOZ localization. To investigate whether the parameter can be used for accurate SOZ localization, the unscented kalman filter (UKF) is employed to estimate the excitatory-inhibitory balance parameter c from the Z6 neural computational model using DRE patients' electrocorticography (ECoG). The results indicate that this parameter follows a unimodal distribution during the pre-ictal period and the post-ictal period, while exhibiting a bimodal distribution during the ictal period. Then, the distribution of this parameter is combined with machine learning methods, and a bagged tree classifier is constructed to localize the SOZ. The classification results demonstrate that the classifier based on parameter distributions exhibits excellent performance, particularly during the post-ictal period, with an average accuracy of 91.60% . Interestingly, SOZ localization is more accurate when no lesions are detected on magnetic resonance imaging (MRI) compared to when lesions are present. Finally, the model parameter distributions of the SOZs are utilized to predict the outcome of epilepsy surgery. Of note, the results demonstrate that the parameter distribution accurately predicts surgical outcomes with an average accuracy of 92.56% . These findings suggest that the distribution of neural computational model parameters may serve as biomarkers for SOZ localization and epilepsy surgery outcome prediction, providing valuable support and assistance for clinical decision-making.

Radar emitter identification (REI) is a crucial function of electronic radar warfare support systems. The challenge emphasizes identifying and locating unique transmitters, avoiding potential threats, and preparing countermeasures. Due to the remarkable effectiveness of deep learning (DL) in uncovering latent features within data and performing classifications, deep neural networks (DNNs) have seen widespread application in REI. In many real-world scenarios, obtaining a large number of annotated radar transmitter samples for training identification models is essential yet challenging. Given the issues of insufficient labeled datasets and abundant unlabeled training datasets, we propose a novel REI method based on a semi-supervised learning (SSL) framework with virtual adversarial training (VAT). Specifically, two objective functions are designed to extract the semantic features of radar signals: computing cross-entropy loss for labeled samples and virtual adversarial training loss for all samples. Additionally, a pseudo-labeling approach is employed for unlabeled samples. The proposed VAT-based SS-REI (SS-VAT) method is evaluated on a radar dataset. Simulation results indicate that the proposed SS-VAT method outperforms the latest SS-REI method in recognition performance.

Epilepsy, a chronic neuropsychiatric brain disorder characterized with recurrent seizures, is closely associated with abnormal neural communications within the brain. Despite that the phase-amplitude coupling (PAC) has been suggested to offer a new way to observe neural interactions during epilepsy, however, few studies pay attention to alterations of the epileptic functional brain network based on PAC, especially on the [Formula: see text] PAC. Therefore, we use scalp electroencephalography (EEG) data of epileptic patients and the [Formula: see text] PAC modulation index (MI) to construct functional brain networks to examine variations of neural interactions during different epileptic phases. Statistically, the findings show that between-channel MI values in the post-ictal period significantly increase compared to that in the pre-ictal period, and the between-channel MI value has a close association with the information of phase and amplitude provided by the channels. Importantly, in both the phase-amplitude and amplitude-phase functional brain networks, the average node degree is remarkably higher in the post-ictal period than that in the pre-ictal period, whereas the characteristic path length in the ictal and post-ictal periods is significantly lower than that in the pre-ictal period. Besides, the average betweenness centrality in the post-ictal period is remarkably higher than that in the ictal period. Interestingly, the positive correlations between within-channel MI values and between-channel MI values can be observed during the pre-ictal, ictal and post-ictal periods. These findings suggest that the [Formula: see text] PAC-based functional brain network may provide a novel perspective to understanding alterations of neural interactions during the epileptic evolution, and may contribute to effectively controlling the spread of epileptic seizures.

Specific emitter identification (SEI) uses advanced techniques to identify radio equipment by analyzing unique characteristics in radio frequency signals. Recently, deep learning (DL) has been considered a promising tools for designing various intelligent SEI methods. This is primarily due to its ability to fully exploit hidden data features and make autonomous classification decisions, leading to effective performance. The existing DL-SEI methods are based on the availability of extensive labeled datasets, however, collecting and annotating such data is challenging and time-consuming in real-world scenarios. Furthermore, these datasets often contain both device-specific and irrelevant features, which limits the adaptability of models to fixed channels. To overcome these challenges, we propose a robust variable-channel SEI (VC-SEI) method. This method uses semantic consistency-powered semi-supervised domain adaptation (SSDA). We introduce domain adversarial training to ensure global semantic consistency (GSC), allowing the extraction of channel-irrelevant features. Additionally, we design two loss functions to maintain local semantic consistency (LSC) for extracting category-relevant features. This approach enables effective domain adaptation. Our SSDA-based VC-SEI method has been rigorously evaluated using the ORACLE RF fingerprinting datasets from 16 USRP X310 radios. When only 1% of training samples in the target domain are labeled, our method achieves 84.20% identification accuracy in the target domain and 92.00% identification accuracy in the source domain. These results surpass those of current state-of-the-art methods. Simulation results confirm the robust identification performance of our proposed VC-SEI method in both source and target domains across all scenarios. Our code can be downloaded fromhttps://github.com/frownean/VC-SEI-based-SSDA.

As agents interact and influence one another in a social network, the opinions they hold about some common topic can change over time. These changes may enable us to infer mechanisms of the network that control how interactions lead to opinion change. Inferring such mechanisms from opinion data could enable analysis of social influence in data-sparse scenarios. However, limited work has focused on this problem, despite its clear value. To address this gap, we create opinion data using agent-based simulation and experimental design. By viewing opinion changes as an information-generating process, opinion dynamics can be studied using entropy. This work explores the relationships between aggregated entropy and five simulation design factors. Three entropy measures are calculated on continuous-valued opinions and are analyzed using a main effects model and cluster analysis. Overall, the choices of influence model and error distribution are most important to the entropy measures, activation regime is important to some measures, and population size is unimportant. Also, design variation can be detected using time-series cluster analysis. These findings may support work in inferring properties about real-world social influence networks using opinion data collected from their members.

Drought is one of the most common natural disasters which may harm ecosystem and economy, and thus, it is important to accurately grasp the change of drought. In this study, an innovative multisource remote sensing drought index improved Temperature-Vegetation-Soil Moisture Dryness Index (iTVMDI) based on passive microwave remote sensing data of FengYun (FY)3B and optical/infrared data is proposed to monitor the drought in Shandong Province, China in 2016. The monitoring results were verified by meteorological data. Results showed that iTVMDI has a negative correlation (R=-0.73) with precipitation and the average correlation coefficient with temperature is 0.76. Overall, iTVMDI can be applied to monitor the temporal and spatial variation of drought conditions.

Drought is one of the major natural disasters which not only causes great damage to ecosystems and the environment, but also seriously affects social and economic activities and residents' lives. In this study, the MODIS- Albedo, LST and NDVI data was used to downscale the FY-3B soil moisture product at 25km to 1km by multivariate statistical regression method. Then the drought of Shandong Province was monitored using the downscaling results in 2016. The results indicate that drought monitoring using multisource remote sensing data is very useful for drought monitoring.

We consider users who may have renewable energy harvesting devices or distributed generators. Such users can behave as consumers or producers (hence, we denote them as prosumers) at different time instances. We consider a tiered market where the grid selects a price function, which reveals price in the real time based on the total demand to the grid. In the real time, a prosumer can buy from another prosumer in an exchange market knowing the price from the grid. The exchange price is set by a platform and can be different for different sellers. A prosumer is a selfish entity, which selects the amount of energy it wants to buy either from the grid or from other prosumers or the amount of excess energy it wants to sell to other prosumers by maximizing its own payoff. However, the strategy and the payoff of a prosumer inherently depend on the strategy of other prosumers as a prosumer can only buy if the other prosumers are willing to sell. We formulate the problem as a coupled constrained game and seek to obtain the generalized Nash equilibrium. We show that the game is a concave potential game and show that there exists a unique generalized Nash equilibrium. We propose a distributed algorithm that converges to the exchange price, which clears the market and achieves the generalized Nash equilibrium. We, finally, show how the grid should select the price function in a day-ahead scenario by computing the estimated demand from the history. Our numerical result shows that the tiered market can reduce the peak load and increase the prosumers' total payoffs.

Soil moisture is an important parameter for the terrestrial water, energy and carbon cycles. Measurement of the dielectric properties is critical for near-surface soil moisture estimation using microwave remote sensing. A number of empirical and semi-empirical models have been derived to describe the relationship. In this study, four widely-used soil dielectric models for soil moisture retrieval, the Wang-Schmugge model, the Dobson model, the Hallikainen model, and the GRMDM (generalized refractive mixing dielectric model) model were compared and the effect of uncertainties of them on soil retrievals were also investigated. Theoretical values of soil dielectric constant were calculated for three soil textures (60% sand, 20% clay; 30% sand, 30% clay; and 20% sand, 60% clay) with the four soil dielectric constant models. The effective soil dielectric constants calculated by models are seen to differ more as the sand content of soil increases. Errors induced by wrong soil texture may exceed 10% for the worst case. The results indicate that the errors induced by the alternative dielectric mixing model may be over the standard accuracy requirement (4% volumetric soil moisture) in soil moisture retrievals.

Response surface methodology (RSM) is a widely used method for simulation optimization. Its strategy is to explore small subregions of the decision space in succession instead of attempting to explore the entire decision space in a single attempt. This method is especially suitable for complex stochastic systems where little knowledge is available. Although RSM is popular in practice, its current applications in simulation optimization treat simulation experiments the same as real experiments. However, the unique properties of simulation experiments make traditional RSM inappropriate in two important aspects: (1) It is not automated; human involvement is required at each step of the search process; (2) RSM is a heuristic procedure without convergence guarantee; the quality of the final solution cannot be quantified. We propose the stochastic trust-region response-surface method (STRONG) for simulation optimization in attempts to solve these problems. STRONG combines RSM with the classic trust-region method developed for deterministic optimization to eliminate the need for human intervention and to achieve the desired convergence properties. The numerical study shows that STRONG can outperform the existing methodologies, especially for problems that have grossly noisy response surfaces, and its computational advantage becomes more obvious when the dimension of the problem increases.

In this paper, we develop a novel statistical detection algorithm following similar principles to that of expectation maximization (EM) algorithm. Our goal is to develop an iterative algorithm for joint channel estimation and data detection in channels that have a long memory and are fast varying in time. At each iteration, starting with an estimate of the channel, we combine a Markov Chain Monte Carlo (MCMC) algorithm for data detection, and an adaptive algorithm for channel tracking, to develop a statistical search procedure that finds joint important samples of possible transmitted data and channel impulse responses. The result of this step, which may be thought as E-step of the proposed algorithm, is used in an M-step that refines the channel estimate, for the next iteration. Excellent behavior of the proposed algorithm is presented by examining it on real data from underwater acoustic communication channels.

Controlled sequential bifurcation (CSB) is a factor-screening method for discrete-event simulations. It combines a multistage hypothesis testing procedure with the original sequential bifurcation procedure to control both the power for detecting important effects at each bifurcation step and the Type I error for each unimportant factor under heterogeneous variance conditions when a main-effects model applies. This paper improves the CSB procedure in two aspects. First, a new fully sequential hypothesis-testing procedure is introduced that greatly improves the efficiency of CSB. Moreover, this paper proposes CSB-X, a more general CSB procedure that has the same error control for screening main effects that CSB does, even when two-factor interactions are present. The performance of the new method is proven and compared with the original CSB procedure.

In this paper, we consider the optimality of the beamforming scheme for both the multiple-input multiple-output (MIMO) point-to-point channel and the MIMO multiple access channel (MAC), where all communication terminals are assumed to be equipped with multiple antennas. For both channels, the channel matrices have correlated elements and are modelled by virtual representation. For the point-to-point channel, i.e., the single user case, we show that the optimal beamforming angle is unique and is independent of the signal-to-noise ratio (SNR). We further show that there exists a certain SNR threshold below which beamforming is optimal and above which beamforming is strictly suboptimal. For the MIMO MAC, we show that to achieve sum capacity, the inputs from different users are independent and their covariance matrices are diagonal. We also derive a necessary and sufficient condition for the optimal input distribution to achieve the sum capacity. Based on these results, we investigate the conditions under which beamforming achieves the sum capacity. We show that the optimal beamforming angles are not unique, and are dependent on both the value of SNR and beamforming angles of other users. We further provide explicit conditions to determine the optimal beamforming angles for a special class of correlated MIMO MACs.