The pursuit of real-time agentic interaction has driven interest in Diffusion-based Large Language Models (dLLMs) as alternatives to autoregressive backbones, promising to break the sequential latency bottleneck.However, does such efficiency gains translate into effective agentic behavior?In this work, we present a comprehensive evaluation of dLLMs (e.g., LLaDA, Dream) across two distinct agentic paradigms: Embodied Agents (requiring longhorizon planning) and Tool-Calling Agents (requiring precise formatting).Contrary to the efficiency hype, our results on Agentboard and BFCL reveal a "bitter lesson": current dLLMs fail to serve as reliable agentic backbones, frequently leading to systematic failure.(1) In Embodied settings, dLLMs suffer repeated attempts, failing to branch under temporal feedback.(2) In Tool-Calling settings, dLLMs fail to maintain symbolic precision (e.g.strict JSON schemas) under diffusion noise.To assess the potential of dLLMs in agentic workflows, we introduce DiffuAgent, a multi-agent evaluation framework that integrates dLLMs as plug-and-play cognitive cores.Our analysis shows that dLLMs are effective in non-causal roles (e.g., memory summarization and tool selection) but require the incorporation of causal, precise, and logically grounded reasoning mechanisms into the denoising process to be viable for agentic tasks.

Large Language Models (LLMs) have shown significant potential as judges for Machine Translation (MT) quality assessment, providing both scores and fine-grained feedback. Although approaches such as GEMBA-MQM have shown state-of-the-art performance on reference-free evaluation, the predicted errors do not align well with those annotated by human, limiting their interpretability as feedback signals. To enhance the quality of error annotations predicted by LLM evaluators, we introduce a universal and training-free framework, MQM-APE, based on the idea of filtering out non-impactful errors by Automatically Post-Editing (APE) the original translation based on each error, leaving only those errors that contribute to quality improvement. Specifically, we prompt the LLM to act as 1) evaluator to provide error annotations, 2) post-editor to determine whether errors impact quality improvement and 3) pairwise quality verifier as the error filter. Experiments show that our approach consistently improves both the reliability and quality of error spans against GEMBA-MQM, across eight LLMs in both high- and low-resource languages. Orthogonal to trained approaches, MQM-APE complements translation-specific evaluators such as Tower, highlighting its broad applicability. Further analysis confirms the effectiveness of each module and offers valuable insights into evaluator design and LLMs selection.

Accurate prediction of future trajectories of multiple aircraft is crucial for decision-making in dynamic game scenarios, but it is highly challenging due to the complex interactions among aircraft and the high dynamics of their movements. Existing models have limitations in capturing both individual characteristics and interactions in multi-trajectory prediction, as graph-based methods rely on predefined structures, social force models confuse individual information, and attention mechanisms continue to struggle with balancing local and global interactions. To address these problems, we propose Agent-SwinTF, a multi-trajectory prediction model based on the 3D Swin Transformer. Firstly, a novel Agent-based Patch method is introduced to maintain the independence of aircraft features within the window and to avoid cross-agent information interference. Secondly, an encoder-decoder framework is constructed using three Trajectory Specific Blocks to model local-global spatio-temporal interactions. Finally, a deep-shallow feature fusion module is built to enhance the representation of complex flight patterns by combining low-dimensional trajectory details from the encoder with high-dimensional semantic information from the decoder. On a self-constructed adversarial game simulation dataset, Agent-SwinTF achieves a Relative Displacement Error Ratio (RDER) value of 1.29, which significantly outperforms the baseline models. Additionally, the average displacement error (ADE) and final displacement error (FDE) are reduced by 52.81% and 25.52%, respectively, in the ablation experiments using the Patch-based approach. Experiments demonstrate that our method achieves superior performance in multi-aircraft trajectory prediction and provides strong support for dynamic game decision-making.

This brief describes a fast finite time adaptive control strategy for p-norm stochastic nonlinear systems (SNSs) using adding a power integrator approach. In terms of convergence speed, finite time control (FTC) offers a substantial advantage in the vicinity of the equilibrium point, but it may exhibit notably slower convergence compared to exponential convergence when the initial state is far from the origin. To overcome this problem, the Lyapunov function is skillfully constructed in this study. A method is devised based on this function, which incorporates the characteristic of adding a power integrator technique and the symbolic function to effectively tackle challenges posed by complex system structures. Then, adaptive control is utilized to deal with the mismatched uncertain nonlinear function. Subsequently, a novel fast FTC strategy is proposed for p-norm SNSs with dead-zone via the back-stepping framework, which ensures the transient performance of the closed-loop system. In comparison to existing results, this controller effectively achieves performance control for p-norm SNSs with dead-zone and mismatched uncertainties functions. Finally, the superiority of the scheme is illustrated by comparative simulations. Note to Practitioners—The FTC problem is a prominent subject in the field of control, playing a crucial role in practical applications. This is especially notable in p-norm nonlinear systems, where the dynamic behavior exhibits uncontrollable linearization characteristics near the origin, introducing complexity to control analysis. Another challenging aspect in nonlinear control is the impact from external disturbances. In practical systems, the presence of random disturbances is inevitable and frequently results in system instability. A fundamental technical barrier is the involvement of Brownian motion integral terms in stochastic nonlinear systems, along with Itô differential, which introduces not only gradients in Lyapunov analysis but also the Hessian of the Lyapunov function. Furthermore, it is observed that mismatched uncertainties and dead-zone phenomena exist in real systems. Designing fast finite-time controllers has become a focal point of controller research. To address these issues, this paper proposes a fast FTC algorithm tailored for a class of p-norm SNSs with mismatched uncertainties. The proposed approach ensures that the system state can converge to the desired region near the origin in an almost fast finite time.

As a critical task in video sequence classification within computer vision, Online Action Detection (OAD) has garnered significant attention. The sensitivity of mainstream OAD models to varying video viewpoints often hampers their generalization when confronted with unseen sources. To address this limitation, we propose a novel Probabilistic Temporal Masked Attention (PTMA) model, which leverages probabilistic modeling to derive latent compressed representations of video frames in a cross-view setting. The PTMA model incorporates a GRU-based temporal masked attention (TMA) cell, which leverages these representations to effectively query the input video sequence, thereby enhancing information interaction and facilitating autoregressive frame-level video analysis. Additionally, multi-view information can be integrated into the probabilistic modeling to facilitate the extraction of view-invariant features. Experiments conducted under three evaluation protocols—cross-subject (cs), cross-view (cv), and cross-subject-view (csv)—demonstrate that the PTMA achieves state-of-the-art performance on the DAHLIA, IKEA ASM, and Breakfast datasets.

In this study, an event-driven fuzzy adaptive control strategy is introduced to address the tracking control problem in a class of stochastic nonstrict-feedback systems under conditions of input delay, unknown control directions and full-state constraints. First, a model for a class of stochastic nonlinear control systems is established based on a ship maneuvering system. Subsequently, the system model is generalized further to obtain more complex nonstrict-feedback systems with unknown control directions. Next, the system is preprocessed to facilitate the design of control strategy. Through the introduction of an intermediate variable, Pade approximation principle is employed and the impact of input delay is eliminated. Subsequently, a nonlinear mapping approach is employed to transform the issue of full-state constraints into an unconstrained problem. This circumvents the complexities caused by the utilization of barrier Lyapunov functions effectively. Furthermore, the principles of fuzzy approximation and adaptive control theory are introduced. Based on the properties of the exponential function, an event-triggered control strategy is designed meticulously to ensure the semiglobal boundedness of all signals in the closed-loop system.

The motivation of this study is to solve the challenges posed by the nonstrict feedback structure and multiple timedelays factors in stochastic systems, which significantly complicate the structure of system structure and make the procedure of controller design more difficult. Firstly, a new LyapunovKrasovskii function is constructed in this study. A method is devised based on this function, which incorporates the characteristic of kernel functions in fuzzy logic systems and the concept of variable separation to effectively tackle challenges posed by complex system structures and time-delay factors, as well as reducing the burden of updating dynamic gains. Additionally, the application of the Nussbaum function technique efficiently resolves the issue of unknown control direction while ingeniously leveraging the distinctive properties of the Nussbaum function and the stochastic Barbalat's lemma. This approach provides a complete theoretical proof of the boundedness of the system signals and guarantees the asymptotically stable in probability. Ultimately, the proposed approach is validated by the exceptional performance of the simulation results for an electromechanical control system.

Redirected walking (RDW) and omnidirectional treadmill (ODT) are two effective solutions to the natural locomotion interface in virtual reality. ODT fully compresses the physical space and can be used as the integration carrier of all kinds of devices. However, the user experience varies in different directions of ODT, and the premise of interaction between users and integrated devices is a good match between virtual and real objects. RDW technology uses visual cues to guide the user's location in physical space. Based on this principle, combining RDW technology with ODT to guide the user's walking direction through visual cues can effectively improve user experience on ODT and make full use of various devices integrated on ODT. This paper explores the novel prospects of combining RDW technology with ODT and formally puts forward the concept of O-RDW (ODT-based RDW). Two baseline algorithms, i.e., OS2MD (ODT-based steer to multi-direction), and OS2MT (ODT-based steer to multi-target), are proposed to combine the merits of both RDW and ODT. With the help of the simulation environment, this paper quantitatively analyzes the applicable scenarios of the two algorithms and the influence of several main factors on the performance. Based on the conclusions of the simulation experiments, the two O-RDW algorithms are successfully applied in the practical application case of multi-target haptic feedback. Combined with the user study, the practicability and effectiveness of O-RDW technology in practical use are further verified.

The pretrained language model (PLM) based metrics have been successfully used in evaluating language generation tasks.Recent studies of the human evaluation community show that considering both major errors (e.g.mistranslated tokens) and minor errors (e.g.imperfections in fluency) can produce high-quality judgments.This inspires us to approach the final goal of the automatic metrics (human-like evaluations) by fine-grained error analysis.In this paper, we argue that the ability to estimate sentence confidence is the tip of the iceberg for PLM-based metrics.And it can be used to refine the generated sentence toward higher confidence and more reference-grounded, where the costs of refining and approaching reference are used to determine the major and minor errors, respectively.To this end, we take BARTScore as the testbed and present an innovative solution to marry the unexploited sentence refining capacity of BARTScore and human-like error analysis, where the final score consists of both the evaluations of major and minor errors.Experiments show that our solution consistently improves BARTScore, outperforming top-scoring metrics in 19/25 test settings.Analyses demonstrate our method robustly and efficiently approaches human-like evaluations, enjoying better interpretability.Our code and scripts will be publicly released in https: //github.com/Coldmist-Lu/ ErrorAnalysis_NLGEvaluation.

Time series forecasting is crucial for many fields, such as disaster warning, weather prediction, and energy consumption. The Transformer-based models are considered to have revolutionized the field of time series. However, the autoregressive form of the Transformer introduces cumulative errors in the inference stage. Furthermore, the complex temporal pattern of the time series leads to an increased difficulty for the models in mining reliable temporal dependencies. In this paper, we propose the Probabilistic Decomposition Transformer model, which provides a flexible framework for hierarchical and decomposable forecasts. The hierarchical mechanism utilizes the forecasting results of Transformer as conditional information for the generative model, performing sequence-level forecasts to approximate the ground truth, which can mitigate the cumulative error of the autoregressive Transformer. In addition, the conditional generative model encodes historical and predictive information into the latent space and reconstructs typical patterns from the latent space, such as seasonality and trend terms. The process provides a flexible framework for the separation of complex patterns through the interaction of information in the latent space. Extensive experiments on several datasets demonstrate the effectiveness and robustness of the model, indicating that it compares favorably with the state-of-the-art.

Electroluminescence imaging becomes a very useful technique to automatically detect defects for solar cells since it can provide high resolution electroluminescence images. However, few methods explicitly consider the visual characteristics of the defects and the noises in solar cells. In this article, a global pairwise similarity and concatenated saliency guided neural network is proposed by fully considering the observed visual characteristics in electroluminescence solar cell images. The proposed network exploits a global pairwise similarity module and a concatenated saliency module to refine the features extracted by the convolutional neural network. The global pairwise similarity module aims to refine the features of an image pixel by modeling long-range dependencies. The concatenated saliency module is exploited to suppress the background and decouple different salient regions to better represent the features of an image. Extensive experiments based on five different baselines, i.e., VGG16, ResNet56, ResNet50, DenseNet40, and GoogleNet, prove that the proposed method significantly outperforms the baseline models and show that both the global similarity module and the concatenated saliency module can help detect defective solar cells in electroluminescence images.

The natural locomotion interface is critical to the development of many VR applications. For household VR applications, there are two basic requirements: natural immersive experience and minimized space occupation. The existing locomotion strategies generally do not simultaneously satisfy these two requirements well. This article presents a novel omnidirectional treadmill (ODT) system named Hex-Core-MK1 (HCMK1). By implementing two kinds of mirror-symmetrical spiral rollers to generate the omnidirectional velocity field, this proposed system is capable of providing real walking experiences with a full-degree of freedom in an area as small as 1.76 m$^{2}$, while delivering great advantages over several existing ODT systems in terms of weight, volume, latency and dynamic performance. Compared with the sizes of Infinadeck and HCP, the two best motor-driven ODTs so far, the 8 cm height of HCMK1 is only 20% of Infinadeck and 50% of HCP. In addition, HCMK1 is a lightweight device weighing only 110 kg, which provides possibilities for further expanding VR scenarios, such as terrain simulation. The system latency of HCMK1 is only 9ms. The experiments show that HCMK1 can deliver a starting acceleration of 16.00 m/s$^{2}$and a braking acceleration of 30.00 m/s$^{2}$.

Locomotion is one of the most important problems in virtual reality. Real walking experience is the key to immersively explore the virtual world. Several strategies have been proposed to solve the problem, but most are not suitable to solve the locomotion problem in Room-Scale VR. The omnidirectional treadmill is an effective way to provide a natural walking experience within the Room-Scale VR. This paper proposes a novel omnidirectional treadmill named HEX-CORE-PROTOTYPE (HCP). The principle of synthesis and decomposition of velocity is applied to form an omnidirectional velocity field. Our system could provide a full degree of freedom and real walking experience in place. Compared to the current best system, the height of HCP is only 40% of it. The application shows the effectiveness of our system to solve the locomotion problem in Room-Scale VR.

As a key component of human-computer intelligent interaction and many real-world applications, the real-time property of facial expression recognition is especially important. However, the recognition result of conventional video-based approaches can not be given until the entire video is finished. In this letter, we deal with early expression detection, which aims to identify the expression as early as possible before its ending. This is a relatively new and challenging problem. Max-margin early event detector (MMED) is a well-known framework, which can make early detection. However, the linearity restricts its applications. We thus introduce kernel learning to model the nonlinear structure of complex data distribution. Moreover, the model is further reformulated in an online setting to address the streaming videos. The high retraining cost and large memory requirement of MMED are thus significantly reduced. In addition, we employ AlexNet architecture to make further comparison with mid-level features. Experiments on two popular video-based expression datasets demonstrate both the effectiveness and efficiency of the proposed method.

Under the noncascade structure, a permanent-magnet synchronous motor (PMSM) regulates the speed and current in one loop. On the one hand, fast dynamic performance requires large transient current to provide a high torque. On the other hand, unlike a cascade control, the q-axis current is no longer governed by a reference current signal. In such a system, the nominal controllers cannot guarantee that the q-axis current is within the required range. But an excessively large transient current may threaten the circuit safety. To solve the overcurrent protection problem, the ordinary solution is to choose conservative control parameters, but the dynamic performance inevitably suffers a certain degree of loss. In order to improve this drawback, a simple and effective control scheme is introduced with a current-constrained technique. By constructing a special nonlinear gain, the punishment mechanism for the q-axis current is established in the control action directly. The proposed control approach does not impose limitations on the initial state. Moreover, it has good robustness to load torque uncertainty and undergoes rigorous theoretical analysis. Besides, the proposed current-constrained controller has a very concise structure. It yields a higher reliability of the PMSM control system. Comparative simulation and experimental results between the classic PID and the current-constrained controller on the PMSM servo system verify the feasibility of the presented control scheme.

The existence of singularities often affects the learning dynamics in feedforward neural networks. In this paper, based on theoretical analysis results, we numerically analyze the learning dynamics of radial basis function (RBF) networks near singularities to understand to what extent singularities influence the learning dynamics. First, we show the explicit expression of the Fisher information matrix for RBF networks. Second, we demonstrate through numerical simulations that the singularities have a significant impact on the learning dynamics of RBF networks. Our results show that overlap singularities mainly have influence on the low dimensional RBF networks and elimination singularities have a more significant impact to the learning processes than overlap singularities in both low and high dimensional RBF networks, whereas the plateau phenomena are mainly caused by the elimination singularities. The results can also be the foundation to investigate the singular learning dynamics in deep feedforward neural networks.

Reliable and accurate short-term wind speed forecasting is of great importance for secure power system operations. In this study, a novel two-step method to construct a multi-model ensemble, which consists of linear regression, multi-layer perceptrons and support vector machines, is proposed. The ensemble members first compete with each other in a number of training rounds, and the one with the best forecasting accuracy in each round is recorded. Then, after all the training rounds, the occurrence frequency of each member is calculated and used as the weight to form the final multi-model ensemble. The effectiveness of the proposed multi-model ensemble has been assessed on the real datasets collected from three wind farms in China. The experimental results indicate that the proposed ensemble is capable of providing better performance than the single predictive models composing it.

Radial basis function (RBF) networks are one of the most widely used models for function approximation and classification. There are many strange behaviors in the learning process of RBF networks, such as slow learning speed and the existence of the plateaus. The natural gradient learning method can overcome these disadvantages effectively. It can accelerate the dynamics of learning and avoid plateaus. In this letter, we assume that the probability density function (pdf) of the input and the activation function are gaussian. First, we introduce natural gradient learning to the RBF networks and give the explicit forms of the Fisher information matrix and its inverse. Second, since it is difficult to calculate the Fisher information matrix and its inverse when the numbers of the hidden units and the dimensions of the input are large, we introduce the adaptive method to the natural gradient learning algorithms. Finally, we give an explicit form of the adaptive natural gradient learning algorithm and compare it to the conventional gradient descent method. Simulations show that the proposed adaptive natural gradient method, which can avoid the plateaus effectively, has a good performance when RBF networks are used for nonlinear functions approximation.