Long-term conversational agents rely on personal memory to maintain coherence and personalization, yet practical systems must operate under context budgets and cope with evolving or contradictory user information. We frame persona memory as a retrieval problem over a growing memory store, and propose REMAP, a reflection-guided memory editing approach for online alignment of persona facts that selectively writes and revises memory entries based on the current dialogue evidence and retrieved related items. The method aims to preserve salient facts while reducing redundancy and resolving apparent conflicts, enabling more efficient context utilization over extended interaction horizons. Experiments on multi-session dialogue datasets show consistent gains in persona-consistent retrieval and response continuity over commonly used memory strategies, while achieving more selective memory updates under comparable operational overhead.

Vehicle trajectory is crucial for autonomous driving. Relatively scattered trajectory data points pose difficulties in modeling the motion’s inherent continuity in spatial and temporal dimensions. Additionally, identifying the driving patterns of vehicles from trajectories is also a significant challenge. These implicit characteristic patterns are difficult to discern from the complex details of the trajectory data. To address these issues, we propose a new framework called Wavelet-Mamba Trajectory Prediction (WMTP), which fuses wavelet analysis through state-space modeling to capture global trends in driving patterns and details of vehicle motion. The approach employs the Discrete Wavelet Transform (DWT) to decompose trajectory data into wavelet coefficients in different time scales and frequencies, and then utilizes these coefficients to generate the trajectory through the Inverse Discrete Wavelet Transform (IDWT). An encoder-decoder neural architecture is proposed for learning potential temporal features from the input trajectory sequences, and these features are projected into the wavelet domain. Wavelet coefficients of future trajectories are generated using different scale-oriented decoders. The estimated coefficients are further used to realize the trajectory prediction via the IDWT module. Experiments demonstrate that WMTP exhibits excellent performance on three large-scale real-world trajectory prediction datasets, with promising robustness and inference speed. The research findings also verify the effectiveness of time - frequency analysis in trajectory prediction tasks.

Goal-conditioned reinforcement learning (GCRL) is an effective method for multi-goal robotic manipulation tasks. Many studies based on hindsight experience replay (HER) and hindsight goal generation (HGG) have achieved the autonomous acquisition of robotic manipulation in reward-sparse environments and have greatly improved the learning efficiency of GCRL. However, these methods perform poorly in environments with obstacles and distant goals. In this paper, we propose hindsight goal diffusion and graph-based experience replay (HGD-GER) for complex robotic manipulation. First, obstacle-avoiding graphs in environments with obstacles are constructed, and the graph-based distance metric between different goals is established. Second, the proposed HGD approach utilizes the inherent denoising mechanism of diffusion models and obstacle-avoiding graph-based distance to generate exploration goals, thereby promoting the exploration of obstacle-bypassing areas. Then, GER module modifies the reward value of experience replay by graph-based distance, thereby avoiding the bias introduced by HER and improving the learning performance of the RL algorithm under sparse reward conditions. Finally, we conducted experiments on three robotic manipulation tasks with obstacles and distant goals, and the results show that the proposed HGD-GER achieves excellent learning performance. Additionally, the proposed method is deployed on the physical robot.

The rapid advancement of urbanization in recent decades has attracted extensive application of cellular automata (CA)-based models to simulate urban growth for planning and decision-making. However, the inaccurate representation of heterogeneous spatial interactions between urban units and the neglect of autocorrelated growth patterns in urbanization lead to unreliable simulation results of CA-based models. To address these two limitations, this study proposes a novel CA-based model integrated with Transformer network and Local Moran’s I, namely TL-CA. The Transformer network is built to quantify heterogeneous interaction between neighbors using the self-attention mechanism. Subsequently, Local Moran’s I is employed to implicitly guide the network in learning spatially autocorrelated patterns of urban growth through auxiliary learning. Finally, the development potential estimated from driving factors, i.e. the network output, is incorporated into CA to simulate urban growth. Land use data from Wuhan (2000–2020) are selected to verify TL-CA’s performance. The results demonstrate that TL-CA achieves the highest simulation accuracy, with an average increase in the figure of merit (FoM) of 9.97%. Attention visualization and residual analysis explain the model’s effectiveness in modeling heterogeneous interactions and autocorrelated growth. Additionally, TL-CA exhibits high computational efficiency and low resource consumption, with sufficient potential to support larger-scale research.

Wheeled robots play a crucial role in Industrial Internet of Things (IIoT)-enabled manufacturing environments, and ensuring their reliable operation is essential for production efficiency and safety. However, their inherent complexity makes them prone to faults, while limited fault data results in imbalanced datasets, posing challenges for deep-learning-based fault diagnosis models. Existing denoising diffusion probabilistic model (DDPM)-based fault diagnosis methods tend to address single-channel scenarios, ignoring the graph relationships inherent in multichannel sensor data. Furthermore, current graph-based DDPM models also struggle in wheeled robot scenarios due to its multimodal nature. To address these challenges, we propose an enhanced DDPM-based method for imbalanced fault diagnosis of wheeled robots. Our method integrates graph operations into the noise prediction network of the DDPM framework, enabling efficient modeling of the complex spatial–temporal relations in multimodal graphical sequence data via a newly designed spatial–temporal graph U-Net (STGU-Net). Additionally, we present a dynamic degradation mechanism for the prior graph, simulating gradual structural changes during the diffusion process. Extensive experiments on a real-world wheeled robot platform demonstrate the superiority of the proposed model over state-of-the-art methods from multiple perspectives, showcasing its effectiveness in generating high-quality data and mitigating the imbalance problem in wheeled robot fault diagnosis.

Hindsight experience replay (HER) has greatly increased the possibility of using deep reinforcement learning (DRL) for robotic manipulation with sparse rewards. However, there are still concerns about low learning efficiency and poor performance due to its insufficient exploration ability and bias against the initial goal introduced by HER. In this article, to solve this problem, a multigoal robotic manipulation DRL method based on adaptive intrinsic curiosity and universal value network fitting (AIC-UVNF) is proposed to further improve the exploration ability and learning performance. Specifically, this method utilizes an improved curiosity mechanism to construct a joint intrinsic reward and adaptively adjust the proportion, which can enhance exploration ability and avoid excessive pursuit of novel states. In addition, a universal value network fitting approach is proposed to incorporate the initial goal into the value function fitting process, which employs the value of the initial goal to eliminate the bias of HER in the algorithm update. Combined with the off-policy soft actor-critic method, AIC-UVNF is verified on multigoal robotic manipulation tasks. The results show that the proposed method achieves better convergence efficiency and learning performance.

We present a real-time, high-accuracy, robust, tightly coupled visual-inertial odometry (VIO) algorithm, including monocular-inertial odometry and stereo-inertial odometry, and uses inertial measurement unit (IMU) pre-integration that is based on fourth-order Runge–Kutta (PK4) and IMU initialization based on maximum a posteriori (MAP) estimation. In particular, we used the multi-state constraint Kalman filter (MSCKF) to fuse vision and IMU measurement data for state estimation. In the optimization stage, we simultaneously considered and optimized all of the historical constraints, and performed multiple iterations to reduce the linearity errors. For further reducing the cumulative error and improving the relocation accuracy, we used a bag-of-words model for global optimization. To lower the computational cost and increase the real-time performance, we set keyframe insertion mechanism and introduced sliding window, and used a new form of Kalman gain that converts the Kalman gain in multi-state constraint Kalman filtering into the inverse of the state dimension. We validated the proposed method by using the EuRoC MAV dataset and KITTI dataset. We performed physics experiments in an outdoor environment with unstable light, to further validate the accuracy and robustness of our method.

Accurate transmission performance prediction of the balise system is important for reliable ground–train communication in high-speed rails. However, the combined effects of ground–train coupling and on-board demodulation make it difficult to predict long-term and volatile holistic transmission performance under multisource disturbances. To address these issues, this article proposes a gate-steered residual interweave architecture. A new gated double-dilated temporal convolutional network (GDTCN) is introduced by combining large and small receptive fields with gating units to extract multiscale features and filter out useless information. It can learn multisource parameter disturbances induced by transient coupling and intermittent demodulation to reduce the prediction errors of occasional volatilities in holistic transmission performance. A new residual interweave GDTCN structure with linear gated cross connections is built to remit the differences between signal and telegram parameter features in view of the inconsistent data transfer modes during coupling and demodulation. This structure can facilitate multipath feature interaction and fusion to enrich discriminative correlation information, thereby improving the prediction accuracy of holistic transmission performance. An adaptive gated attention mechanism is designed to exploit correlation and dependent information between coupling and demodulation in a weighted fusion manner. It can minimize prediction error accumulation to enhance long-term holistic transmission performance forecasting. Extensive experiments demonstrate that the proposed architecture can perform predictions with high accuracy and efficiency under different rail conditions. The proposed method can provide early warning before ground balise or on-board module failures to improve the reliability and availability of ground–train communication.

Efficient point cloud visualization is indispensable for practical applications. In the context of point cloud visualization, 3-D rendering can be viewed as the kernel that transforms 3-D points into a 2-D scene image. Compared with traditional point-based rendering (PBR), neural image-based rendering (NIBR) has gradually emerged as a feasible solution for point cloud rendering. To efficiently render sparse and colorless mobile laser scanning (MLS) point cloud, we propose a novel neural rendering framework based on deep generative learning, named MLS point cloud rendering with generative adversarial network (MPR-GAN). In this framework, perspective projection with intrinsic parameter scaling and cumulative distribution normalization is first utilized to transform the 3-D point cloud into a compact 2-D image; a conditional generative adversarial network (CGAN)-based rendering model is then proposed to generate a photorealistic scene image from the projected 2-D image. In this CGAN model, the asymmetric encoder–decoder generator can implement inpainting and true colorization using context feature capturing and edge information perception; a multiscale discriminator is built to guarantee the model output with global consistency and local details. Moreover, a hybrid loss function is designed to improve the visual quality of the generated images with similarity constraints from both the content and the structure. Two public MLS point cloud datasets are selected and employed to carry out extensive evaluation using MPR-GAN and other baseline frameworks. The experimental results demonstrate that MPR-GAN achieves the state-of-the-art rendering performance in terms of peak signal-to-noise ratio (PSNR) and structural similarity (SSIM). Furthermore, the efficiency analysis shows that MPR-GAN can support real-time rendering, achieving end-to-end rendering from raw points.

Event sequence mining is often used to summarize patterns from hundreds of sequences but faces special challenges when handling racket sports data. In racket sports (e.g., tennis and badminton), a player hitting the ball is considered a multivariate event consisting of multiple attributes (e.g., hit technique and ball position). A rally (i.e., a series of consecutive hits beginning with one player serving the ball and ending with one player winning a point) thereby can be viewed as a multivariate event sequence. Mining frequent patterns and depicting how patterns change over time is instructive and meaningful to players who want to learn more short-term competitive strategies (i.e., tactics) that encompass multiple hits. However, players in racket sports usually change their tactics rapidly according to the opponent's reaction, resulting in ever-changing tactic progression. In this work, we introduce a tailored visualization system built on a novel multivariate sequence pattern mining algorithm to facilitate explorative identification and analysis of various tactics and tactic progression. The algorithm can mine multiple non-overlapping multivariate patterns from hundreds of sequences effectively. Based on the mined results, we propose a glyph-based Sankey diagram to visualize the ever-changing tactic progression and support interactive data exploration. Through two case studies with four domain experts in tennis and badminton, we demonstrate that our system can effectively obtain insights about tactic progression in most racket sports. We further discuss the strengths and the limitations of our system based on domain experts' feedback.

Seasonal time series Forecasting remains a challenging problem due to the long-term dependency from seasonality. In this paper, we propose a two-stage framework to forecast univariate seasonal time series. The first stage explicitly learns the long-range time series structure in a time window beyond the forecast horizon. By incorporating the learned long-range structure, the second stage can enhance the prediction accuracy in the forecast horizon. In both stages, we integrate the auto-regressive model with neural networks to capture both linear and non-linear characteristics in time series. Our framework achieves state-of-the-art performance on M4 Competition Hourly datasets. In particular, we show that incorporating the intermediate results generated in the first stage to existing forecast models can effectively enhance their prediction performance.

There have been several attempts to integrate a spatial visual attention mechanism into an image caption model and introduce semantic concepts as the guidance of image caption generation. High-level semantic information consists of the abstractedness and generality indication of an image, which is beneficial to improve the model performance. However, the high-level information is always static representation without considering the salient elements. Therefore, a semantic attention mechanism is used for the high-level information instead of conventional of static representation in this article. The salient high-level semantic information can be considered as redundant semantic information for image caption generation. Additionally, the generation of visual words and non-visual words can be separated, and an adaptive attention mechanism is employed to realize the guidance information of image caption generation switching between new fusion information (fusion of image feature and high-level semantics) and a language model. Therefore, visual words can be generated according to the image features and high-level semantic information, and non-visual words can be predicted by the language model. The semantics attention, adaptive attention, and previous generated words are fused to construct a special attention module for the input and output of long short-term memory. An image caption can be generated as a concise sentence on the basis of accurately grasping the rich content of the image. The experimental results show that the performance of the proposed model is promising for the evaluation metrics, and the captions can achieve logical and rich descriptions.

Deep neural network (DNN) has many advantages. Autonomous driving has become a popular topic now. In this paper, an improved stack autoencoder based on the deep learning techniques is proposed to learn the driving characteristics of an autonomous car. These techniques realize the input data adjustment and solving diffusion gradient problem. A Raspberry Pi and a camera module are mounted on the top of the car. The camera module provides the images needed for training the DNN. There are two stages in the training. In the pre-training process, an improved autoencoder is trained by the unsupervised learning mechanism, and the characterization of the track is extracted. In the fine-tuning stage, the whole network is trained according to the labeled data, and then this model learns the driving characteristics better according to the samples. In the experimental stage, the car will predict the action of the car by the trained model in the autonomous mode. The experiment exhibits the effectiveness of the proposed model. Compared with the traditional neural network, the improved stack autoencoder has a better generalization ability and faster convergence speed.

This paper reports a new general framework of focused web crawling based on "relational subgroup discovery". Predicates are used explicitly to represent the relevance clues of those unvisited pages in the crawl frontier, and then first-order classification rules are induced using subgroup discovery technique. The learned relational rules with sufficient support and confidence will guide the crawling process afterwards. We present the many interesting features of our proposed first-order focused crawler, together with preliminary promising experimental results.

Link context has been exploited extensively ever since the advent of the World Wide Web, but the approach to extracting precise link context has not been fully explored and many state-of-the-art extraction methods are based on simplistic heuristics and require ad-hoc parameters. In this paper, we propose a novel two-step extraction model, which aims to systematically derive link context of quality as high as anchor text. In the macroscopic analysis step, a systematic web page structure analysis is performed to locate the content cohesive text region and potential relevant header or header like tags. In the microscopic extraction step, an English parser is used to extract the relevant sentence fragments in the text region and the nearest heading text is encompassed if the need arises. Preliminary experimental results proved our approach's effectiveness.