Continuous sign language recognition (CSLR) technology enables social communication for the hearing-impaired by converting sign language videos into text. However, due to the limited receptive fields of convolutional networks and inefficient long-range dependency modeling in temporal modules, current methods find it difficult to capture cross-regional and high-order dynamic semantics in complex gestures. To address these limitations, we propose a dynamic spatiotemporal hypergraph network named HyperSign, which optimizes feature learning through innovative graph architectures. For single-frame spatial modeling, we propose a saliency-aware spatial graph construction strategy that dynamically quantifies semantic saliency by integrating feature complexity and motion intensity information from patches. This strategy can adaptively adjust node connectivity based on the computed saliency, thereby enabling the graph structure to focus on information-dense regions such as hands and faces. For temporal dependency modeling, we abandon the conventional pairwise frame interactions and propose a temporal hypergraph construction method. This method employs a learnable clustering algorithm to aggregate semantically correlated nodes within temporal windows into hyperedges, thereby explicitly capturing high-order associations within individual gesture actions that span multiple frames. Extensive experiments on the PHOENIX14, PHOENIX14-T, and CSL-Daily datasets demonstrate that HyperSign outperforms the state-of-the-art (SOTA) approaches in CSLR without any additional annotation information, establishing a new feature learning paradigm for the CSLR task.

Conversational recommender systems (CRSs) aim to mine user preferences and recommend appropriate items through natural language dialogue. A complete CRS typically consists of a recommendation module and a conversation module, which generate high-quality recommended items and fluent natural language responses, respectively. Existing research usually constructs and trains the two modules separately, leading to inconsistencies in input information and construction methods for different subtasks. To address the limitations of the above methods, we propose a joint training framework based on knowledge-aware prompt tuning to build an integrated conversational recommender system (KPICRS). First, we adopt contrastive learning to align the semantic space of the embeddings of the conversation context and the knowledge graph entity, and then generate a prompt template as the unified input of the joint training framework through the prompt encoder. Specifically, we incorporate augmented similar user representations into the prompt template, which helps to alleviate the data sparsity and cold-start problems. Next, we jointly train the pretrained language model (PLM) and the prompt encoder so that the PLM can simultaneously generate predictions for both conversation subtask and recommendation subtask. Extensive experiments on two public English and Chinese CRS datasets demonstrate that our model achieves highly competitive performance.

Retrieval-augmented generation (RAG) has been proposed to mitigate the hallucination problem of large language models (LLMs) in knowledge-intensive tasks by incorporating external knowledge. However, in multi-hop question answering, existing iterative retrieval methods often struggle to maintain focus on key information. As the number of retrieval iterations increases, the generated queries can gradually drift from the correct reasoning path, and irrelevant or noisy information may accumulate, ultimately reducing reasoning accuracy. To address these challenges, we propose a novel retrieval-augmented generation method for multi-hop question answering based on structured planning. First, our approach employs pre-retrieval question planning to provide semantic guidance for iterative retrieval, ensuring greater consistency between retrieval and reasoning. In addition, we introduce a structured evidence extraction mechanism to effectively filter out noise in the retrieved information, leading to improved reasoning accuracy. Experimental results on three open-domain multi-hop question answering datasets demonstrate that our method can effectively alleviate the impact of retrieval bias and retrieval noise and exhibit competitive performance.

In continuous sign language recognition (CSLR), 2D-CNN-based extractors are often insufficiently trained for spatial capture and struggle with temporal modeling. This leads to incomplete spatial discrimination, hindering the understanding actions across frames. To address these limitations, we propose Hierarchical Spatial-Temporal Enhancement network (HSTE) through two key modules: Cross-scale Semantic Alignment (CSA) and Temporal Extension Shift (TES). CSA innovatively utilizes multi-scale features generated within the network, enriching feature representation through semantic alignment across scales. By integrating a novel temporal shift strategy with dilated convolutions, TES expands the receptive field and captures temporal changes between frames. These modules work independently and are hierarchically integrated into the network in a plug-and-play manner. Extensive experiments show that our method achieves state-of-the-art performance on the challenging CSLR benchmarks: PHOENIX14, PHOENIX14-T, and CSL-Daily. Code will be available at https://github.com/justlis/HSTENet.

Current continuous sign language recognition (CSLR) methods typically rely on single or adjacent frames for calculations, which can overlook broader contextual information and result in lower accuracy. To address this issue, we introduce CVSign, which constructs an extended contextual space frame by frame while enabling comprehensive cross-frame interaction. Specifically, we present two innovative modules: Contextual Correspondence Awareness (CCA) and Contextual Variability Awareness (CVA). CCA enhances the relevance of contextual features by utilizing cross-frame multi-head query attention to identify and prioritize related areas while suppressing irrelevant regions. CVA captures motion changes at varying speeds by employing difference calculations between multiple frames, effectively minimizing static redundancy. Remarkably, experimental results show that CVSign outperforms the previous state-of-the-art method by a clear margin on widely used datasets, including PHOENIX14, PHOENIX14-T, and CSL-Daily.

Multibehavior collaborative filtering recommendations can significantly alleviate data sparsity issues caused by insufficient single-behavior information, enhancing recommendation performance. However, current multibehavior recommendation methods simply concatenate different behavior representations without further exploring the interactive information between behaviors, thus limiting recommendation effectiveness. To address the limitations, we propose the graph contrastive learning for multibehavior recommendation (GCMR) model. First, we use a shared bottom to capture the connections between different behaviors of each user. Then, we introduce a GCN-based multibehavior contrastive learning approach that employs cross-layer and cross-behavior contrastive learning to capture intrabehavior and cross-behavior network interaction information, which enhances user and item representations. Additionally, we propose a multibehavior feature fusion strategy that integrates user representations (and item representations) to fully exploit latent information of different behaviors and improve network representation performance. Extensive experiments on three open-source datasets demonstrate that the GCMR model outperforms the state-of-the-art, especially on the Tmall dataset, where GCMR achieved an improvement of 19.10% in HR@10 and 16.50% in NDCG@10 over the best baseline.

To accelerate magnetic resonance imaging (MRI), image reconstruction from under-sampled measurements has been widely used. Recently, the convolutional-Transformer hybrid architecture has dominated the field of MRI reconstruction. To improve calculation performance, two multi-scale (MS) strategies are usually adopted: the one imposed in the intra-cascades in a U-shape style and the one lying in the inter-cascades in a pyramid manner. The two MS strategies have their own benefits but have not been combined together for boosting performance. In this work, we proposed a dynamic Hybrid Unrolled Multi-Scale Network (dHUMUS-Net) by incorporating the two MS strategies together. A novel Optimal Scale Prediction Network is presented to dynamically estimate the optimal scales for all cascades of dHUMUS-Net. Experiments on the fastMRI dataset demonstrate the effectiveness of our method over the state-of-the-art methods.

GNN-based social recommendation aims to use social network information to improve recommendation performance of traditional user–item interaction network (U–I network). However, in graph neural network information aggregation, both social networks and U–I networks inevitably have noise, which affects accuracy of recommendation results. To reduce the noise impact of network data, we propose Robust Social Recommendation based on Contrastive Learning and Dual-Stage Graph Neural Network (CLDS). First, considering instability of social networks, we propose the social preference network. It is robust and retains only social friend relationships with common preferences. Based on it and U–I network, we construct a social recommendation pre-training model. Next, we propose self-contrastive learning method. The method initializes multiple social network node representations through Gaussian distribution , pre-training and random disturbance, respectively. Then, it uses contrastive learning on the generated multiple node representations to enhance the robustness of node representation. Finally, CLDS avoids directly capturing potentially user–user and item–item information in U–I networks which is incomplete and untrusted. And instead, it only extracts user–item information to reduce the noise generated by GNN-based U–I network information aggregation. We conduct experiments under the open-source real network dataset. The experimental results show that CLDS outperforms state-of-art methods in social recommendation. The code is available at: https://github.com/Andrewsama/CLDS-master .

In recent years, precision medicine has been consistently studied and employed in cancer treatment. One of the main challenges in precision medicine is accurately predicting a cancer patient’s response to a specific drug(s) using computational models. Due to the heterogeneity of cancer, the distribution between the drug response profiles based on cell lines and real samples is often different, which leads to out-of-data distribution problems that affect putting the existing predictive models into practical application. To overcome this limitation, a novel framework to predict cancer drug response based on transfer learning from combined genomic profiles and chemical structure embedding of drugs is proposed in this paper. This framework features a denoising auto-encoder (DAE) module designed to remove noise from genomics profiles and a graph convolutional network (GCN) employed to extract the embedding of pharmaceutical chemical structures. Through the experiments, our model outperforms recently published methods on Genomics of Drug Sensitivity in Cancer (GDSC) datasets, and gives promising results on personalized drug response prediction on TCGA pan-cancer datasets, especially for the response of two commonly used anti-cancer drugs―Gemcitabine and Cisplatin. In summary, our study designed a computational framework with the potential capability of predicting personalized cancer drug response, which has been preliminarily validated for its effectiveness.

Accurate and effective localization of reliability-critical gates (RCGs) is one of the important prerequisites for low-cost circuit fault tolerance in the early stages of circuit design. This article introduces an accurate and effective approach for localizing RCGs in combinational logic blocks through a benchmarking technique. In the proposed approach, uniform non-Bernoulli sequences are used to produce a set of input vectors for driving circuits. A full-period linear congruential algorithm is employed to generate a sequence that provides the sampled order for the RCGs to be analyzed. This ensures that each gate in the circuit is treated as fairly as possible. To accelerate the localization process, an input-vector-based pruning technique combined with a counting method is also introduced to identify the specified number of RCGs. Then, the criticality of gate reliability for each RCG is measured through benchmarking. A clustering algorithm carries out the convergence checking for the proposed approach. The performance of the proposed approach was evaluated in terms of accuracy, stability, and time-space overhead by various simulations on 74-series circuits and ISCAS-85 benchmark circuits. The results show that its accuracy is close to that of the Monte Carlo model and its stability is better than that of other approximate methods. Moreover, compared with approximate methods, the time overhead of our approach is advantageous in the presence of similar memory overheads.

This paper proposes a novel local stereo matching approach based on self-adapting matching window. We improve the accuracy of stereo matching in 3 steps. First, we integrate shape and size information, and construct robust minimum matching windows by applying a self-adapting method. Then, two matching cost optimization strategies are employed for handling both occlusion regions and image borders. Last, we perform a refinement algorithm for obtaining more accurate depth map. Experiment results on the Middlebury stereo image pairs prove that the proposed matching method performs equally well in comparison with other state-of-the-art local approaches.

To compensate for the shortcomings of existing methods used in TSP (Traveling Salesman Problem), such as the accuracy of solutions and the scale of problems, this paper proposed an improved particle swarm optimization by using a self-organizing construction mechanism and dynamic programming algorithm. Particles are connected in way of scale-free fully informed network topology map. Then dynamic programming algorithm is applied to realize the evolution and information exchange of particles. Simulation results show that the proposed method with good stability can effectively reduce the error rate and improve the solution precision while maintaining a low computational complexity.

This paper proposed a hybrid modeling method based on mechanism analysis, identification and RBF neural networks. First, Get a industrial object's low-order model by the mechanism analysis and identification method. Second, adopt RBF neural networks modeling method to compensate unmodeled high-order model. The sum of the low-order model and high-order model is the hybrid model. This kind of hybrid model has more accuracy than a model which is gotten by mechanism analysis and identification method and has more generalization capability than a model which is gotten by neural networks modeling method.