Conductor avatar plays a dual role in immersive Virtual Reality (VR) interactive systems by interpreting musical scores and guiding orchestral performance. Rule-based score-driven methods ensure precise synchronization with predefined conducting templates or videos, but are constrained by pre-authored data. Audio-driven frameworks offer greater adaptability through real-time gesture generation but often fail to capture the symbolic semantics of musical scores. To overcome these limitations, we propose a novel score-driven gesture generation framework that translates symbolic musical representations into plausible conducting gestures. Our approach adopts a two-stage architecture, combining a comparative learning stage for pre-training a score encoder with a generative learning stage for gesture synthesis. The score encoder explicitly models musical features such as tempo, chord, intensity, and cycle semantics, directly informing gesture generation. To support this research, we introduce Multimodal Symphonic Conducting Dataset (MSCD), the first synchronized dataset comprising conducting gestures, performance audio, and editable symbolic scores, effectively bridging the gap between musical semantics and gesture synthesis. Qualitative and quantitative analyses are provided to demonstrate the effectiveness of our approach, while a user study is designed to identify the strengths and limitations of the current work.

In educators' ongoing pursuit of enriching remote education, Virtual Reality (VR)-based telepresence has shown significant promise due to its immersive and interactive nature. Existing approaches often rely on point cloud or NeRF-based techniques to deliver realistic representations of teachers and classrooms to remote students. However, achieving low latency is non-trivial, and maintaining high-fidelity rendering under such constraints poses an even greater challenge. This paper introduces LiteAT, a data-lightweight and user-adaptive VR telepresence system, to enable real-time, immersive learning experiences. LiteAT employs a Gaussian Splatting-based reconstruction pipeline that integrates an SMPL-X-driven dynamic human model with a static classroom, supporting lightweight data transmission and high-quality rendering. To enable efficient and personalized exploration in the virtual classroom, we propose a user-adaptive viewpoint recommendation framework that dynamically suggests high-quality viewpoints tailored to user preferences. Candidate viewpoints are evaluated based on multiple visual quality factors and are continuously optimized based on recent user behavior and scene dynamics. Quantitative experiments and user studies validate the effectiveness of LiteAT across multiple evaluation metrics. LiteAT establishes a versatile and scalable foundation for immersive telepresence, potentially supporting real-time scenarios such as procedural teaching, multimodal instruction, and collaborative learning.

In humanity’s ongoing quest to craft natural and realistic avatars within virtual environments, the generation of authentic eye gaze behaviors stands paramount. Eye gaze not only serves as a primary non-verbal communication cue, but it also reflects cognitive processes, intent, and attentiveness, making it a crucial element in ensuring immersive interactions. However, automatically generating these intricate gaze behaviors presents significant challenges. Traditional methods can be both time-consuming and lack the precision to align gaze behaviors with the intricate nuances of the environment in which the avatar resides. To overcome these challenges, we introduce a novel two-stage approach to generate context-aware head-and-eye motions across diverse scenes. By harnessing the capabilities of advanced diffusion models, our approach adeptly produces contextually appropriate eye gaze points, further leading to the generation of natural head-and-eye movements. Utilizing Head-Mounted Display (HMD) eye-tracking technology, we also present a comprehensive dataset, which captures human eye gaze behaviors in tandem with associated scene features. We show that our approach consistently delivers intuitive and lifelike head-and-eye motions and demonstrates superior performance in terms of motion fluidity, alignment with contextual cues, and overall user satisfaction.

In recent years, artificial intelligence technologies have been successfully applied in time series prediction and analytic tasks. At the same time, a lot of attention has been paid to financial time series prediction, which targets the development of novel deep learning models or optimize the forecasting results. To optimize the accuracy of stock price prediction, in this paper, we propose a clustering-enhanced deep learning framework to predict stock prices with three matured deep learning forecasting models, such as Long Short-Term Memory (LSTM), Recurrent Neural Network (RNN) and Gated Recurrent Unit (GRU). The proposed framework considers the clustering as the forecasting pre-processing, which can improve the quality of the training models. To achieve the effective clustering, we propose a new similarity measure, called Logistic Weighted Dynamic Time Warping (LWDTW), by extending a Weighted Dynamic Time Warping (WDTW) method to capture the relative importance of return observations when calculating distance matrices. Especially, based on the empirical distributions of stock returns, the cost weight function of WDTW is modified with logistic probability density distribution function. In addition, we further implement the clustering-based forecasting framework with the above three deep learning models. Finally, extensive experiments on daily US stock price data sets show that our framework has achieved excellent forecasting performance with overall best results for the combination of Logistic WDTW clustering and LSTM model using 5 different evaluation metrics.

Pre-Trained Models (PTMs) can learn general knowledge representations and perform well in Natural Language Processing (NLP) tasks. For the Chinese language, several PTMs are developed, however, most existing methods concentrate on modern Chinese and are not ideal for processing classical Chinese due to the differences in grammars and semantics between these two forms. In this paper, in order to process two forms of Chinese uniformly, we propose a novel Classical and Modern Chinese pre-trained language model (CANCN-BERT), with the advantage of effectively processing both classical and modern Chinese, which is an extension of BERT. Form-aware pre-training tasks are elaborately designed to train our model, so as to better adapt it to classical and modern Chinese corpus. Moreover, we define a joint model, proposing dedicated optimization methods through different paths with the control of the switch mechanism. Our model merges characteristics of both classical and modern Chinese, which can adequately and efficiently enhance the representation ability for both forms. Extensive experiments show that our model outperforms baseline models on processing classical and modern Chinese and achieves significant and consistent improvements. Also, the results of ablation experiments demonstrate the effectiveness of each module.

Knowledge Graph (KG) representation learning aims to encode both entities and relations into a continuous low-dimensional vector space. Most existing methods only concentrate on learning representations from structural triples in Euclidean space, which cannot well exploit the rich semantic information with hierarchical structure in KGs. In this paper, we propose a novel DataType-aware hyperbolic knowledge representation learning model called DT-GCN, which has the advantage of fully embedding attribute values of data types information. We refine data types into five primitive modalities, including integer, double, Boolean, temporal, and textual. For each modality, an encoder is specifically designed to learn its embedding. In addition, we define a unified space based on Euclidean, spherical, and hyperbolic space, which is a continuous curvature space that combines advantages of three different spaces. Extensive experiments on both synthetic and real-world datasets show that our model is consistently better than the state-of-the-art models. The average performance is improved by 2.19% and 3.46% than the optimal baseline model on node classification and link prediction tasks, respectively. The results of ablation experiments demonstrate the advantages of embedding data types information and leveraging the unified space.

Random walk is an effective network-based method for information mining. However, the computational complexity of random walk limits its application on large scale datasets. This study evaluating accuracy and performance of graphics processing units (GPUs) in accelerating random walk computation on heterogeneous networks. We first introduce a general heterogeneous network to describe user-item relevance model. We apply GPU-accelerated random walk method to this network and investigate the performance of the method. We demonstrate our method via large-scale experiments across MovieLens datasets and obtain a more than 12× speedup. The results shows the effectiveness and promise of the approach.