Search

Most existing aspect sentiment triplet extraction models emphasize the adoption of novel tagging scheme to jointly extract three elements of sentiment triplets, but they overlook the intrinsic information of individual words, including the implicit relationships between words, which results in the inaccurate triplet extraction. In this paper, we propose a novel method named Affective Commonsense Knowledge Enhanced Dependency Graph, which considers the sentiment information contained in each word and the relative positional distance relationships between them. Specifically, a dependency graph is generated, incorporating the sentiment information obtained through an external knowledge base called SenticNet. This dependency graph is integrated into a multi-layer graph convolutional network to enhance the sentiment dependency relationships between words. Additionally, an attention mechanism with relative position embeddings is utilized to acquire word representations which incorporate contextual and syntactic information. Furthermore, an Expanded Grid Tagging Scheme is employed to provide more accurate description of the relationships between words. In terms of the F1 metric, on the four SemEval datasets, 14res, 14lap, 15res, and 16res, our model improvements were 1.64, 1.66, 2.01, and 2.56 when compared to the baselines, respectively. On the four datasets of ASTE-Data-V2, our model improvements were 0.41, 1.03, 1.12, and 0.34, respectively. Experimental results demonstrate the superior performance of our method. [ABSTRACT FROM AUTHOR]

Pretrained models in abstractive summarization enable the summarization model to generate coherent summaries flexibly, but it also introduces unfactual hallucinations when generating novel words. To circumvent this drawback, we propose A Guided Tensor Product framework for abstractive Summarization (GTPSum). GTPSum utilizes highlighted sentence guidance signal and role embedding in the form of guided tensor product representations (GTPRs) to enhance the generation capacity for short document text. The guidance signal assists the model in focusing on crucial information, while the role embedding aids GTPSum in capturing the syntactic structure relationships present in the sentences. By employing this approach, the framework can not only focus on the essential information from the source document but also grasp the inherent sentence structure, resulting in generated summaries that closely align with the original content. The experimental results demonstrate the strong performance of our approach on the CNN/Daily Mail (44.28/20.88/41.08 ROUGE-1/-2/-L) and performed well on the other three datasets. Moreover, the comparison between our model generated summaries and those generated by the baseline models through human evaluation shows that our summaries contain more critical information, have better readability, and still maintain succinctness. [ABSTRACT FROM AUTHOR]

Summary-level extractive summarization is often regarded as a text-matching task, which selects the summary that is semantically closest to the source document by a matching model. However, the method tends to select candidate summaries with more sentences, because it calculates the semantic similarity between the candidate summaries and the source document. To address the issue, we propose a novel set-based summary ranking strategy for extractive summarization called SeburSum, which selects a summary by examining its semantic similarity with mutually exclusive candidate summaries, rather than its similarity to the source document. In contrast to conventional extractive summarization methods that rely on well-trained extractive models trained on labeled data to select sentences, the ranking strategy eliminates the need for labeled data and enhances the versatility of both supervised and unsupervised extractive summarization. To improve the calculation accuracy of semantic similarity between candidates, we construct a contrastive learning framework with a task-specific contrastive loss to learn vector representations for each candidate summary. Experimental results show that in the supervised extractive summarization, we achieve state-of-the-art extractive performance on the CNN/Daily Mail, Reddit, and Xsum with ROUGE-1 scores of 45.49, 26.71, and 25.77, respectively, and outperform the previous summary-level baseline by 1.08, 1.62, and 0.91. In the unsupervised extractive summarization, we achieved state-of-the-art performance on the CNN/Daily Mail dataset with 42.97, 20.14 and 39.14 for ROUGE-1, ROUGE-2 and ROUGE-L, respectively, outperforming the latest state-of-the-art results by 1.71, 1.96, and 1.93. [ABSTRACT FROM AUTHOR]

Knowledge base question answering aims to answer a question over a knowledge base. Multi-hop knowledge base question answering is a challenging task because it requires multi-step reasoning according to the question to get the answer. Existing models infer the answer over a static subgraph or attend to different parts of the question through an intermediate signal. The former obtains limited semantic information, and the latter provides limited reasoning due to the weak supervision signal. In this paper, we eliminate the limitation of static subgraph reasoning by dynamically expanding subgraphs, which connect the question and subgraph to form a joint subgraph. We then adjust the dynamic subgraph to enable reasoning at each step. Specifically, at each step, the question connects different subgraphs, respects the context while paying attention to a specific part of the question, generates a strong intermediate signal, acts on the subsequent reasoning, and finally obtains a correct answer. A large number of experiments on three datasets show that our method performs better than previous state-of-the-art models. [ABSTRACT FROM AUTHOR]

Knowledge-based visual question answering aims to associate external knowledge facts for answering questions about images. Most existing methods emphasize high-order associations between knowledge facts and questions, and fail to consider the negative effects of unnecessary knowledge facts in multi-hop reasoning. In this paper, we propose a D ual- S tream A ttention M ulti-hop R easoning (DSAMR) architecture that constructs two different attention streams to mitigate unnecessary knowledge facts. This dual-stream mechanism enables the model to reduce the attention weights on unnecessary knowledge while gathering essential knowledge by learning the implicit correlations between knowledge facts and questions. In addition, we designed a hypergraph knowledge extraction module in the architecture to extract optimal knowledge facts by evaluating the relevance of each knowledge fact to the question. The experimental results demonstrate the effectiveness of our method not only on the knowledge-based visual question answering dataset KVQA, but also on the multi-hop question answering dataset PathQuestion. • A dual-stream attention multi-hop reasoning architecture to enhances model's reasoning ability. • A hypergraph knowledge extraction method is proposed. • The proposed model achieves competitive results compared to latest baselines. [ABSTRACT FROM AUTHOR]

Extractive document summarization (EDS) is usually seen as a sequence labeling task, which extracts sentences from a document one by one to form a summary. However, extracting sentences separately ignores the relationship between the sentences and documents. One solution is to use sentence position information to enhance sentence representation, but this will cause the sentence-leading bias problem, especially in news datasets. In this paper, we propose a novel sentence centrality for the EDS task to address these two problems. The sentence centrality is based on directed graphs, while reflecting the sentence-document relationship, it also reflects the sentence position information in the document. We implicitly strengthen the relevance of sentences and documents by using sentence centrality to enhance sentence representation. Notably, we replaced the sentence position information with sentence centrality to reduce sentence-leading bias without causing model performance degradation. Experiments on the CNN/Daily Mail dataset showed that EDS models with sentence centrality significantly improved compared with baseline models. [ABSTRACT FROM AUTHOR]

Currently, many summarization models are based on Transformer architectures, which regard text as a connected graph of words and tokens, thus fail to fully leverage the text's hierarchical structure. However, hierarchical structural information is crucial in understanding long documents, especially scientific papers. To exploit the structural information, we propose a H eterogeneous T ree structure augmented with T riplet P ositions for extractive Sum marization of scientific papers (HTPosum). The proposed model constructs a heterogeneous tree for each document, introducing section nodes and a global node to capture the documents' structural information. The tree is then augmented with a novel triplet position, which consists of three values that represent the depth, parent width position, and sibling width position of each node in a tree. The triplet position provides a unified view of the document structure. Experimental results show that the proposed model achieves state-of-the-art extractive performance on the PubMed and arXiv datasets compared with the latest scientific paper summarization models. • A heterogeneous tree is built to model scientific papers' hierarchical structure. • A triplet position and unique triplet position encoding method is proposed. • The proposed model achieves competitive results compared to latest baselines. [ABSTRACT FROM AUTHOR]

In recent years, personalized dialogue systems have drawn growing attention as they can serve as a personalized assistant for a specific user. A key feature of personalized dialogue systems is the ability to maintain the consistent personality and contextual expression that matches the user's individual style and needs. Numerous approaches have been developed for creating personalized dialogue systems. However, most existing approaches depend heavily on explicit persona information, such as predefined persona-defining sentences or key-value persona data. When using predefined explicit user persona knowledge in real-world situations, users need to spend more time crafting clear and descriptive persona information. To address this issue, we introduce IMPACT, a generative-based personalized dialogue model that leverages dialogue history to discover implicit persona knowledge. Both implicit and explicit persona knowledge are equally valuable. IMPACT is designed to discover the user's personalized expression style within dialogues and generate responses that reflect the consistent style. Achieving dialogue consistency requires modeling the complex relationships between the user's message, persona, and dialogue history. In addition, we use an unlikelihood training objective to optimize the model and further improve dialogue consistency. We conduct comprehensive experiments on two large datasets, and the results show that our method outperforms all baseline models. [ABSTRACT FROM AUTHOR]

The pursuit of higher recognition accuracy and speed with smaller model sizes has been a major research topic in the detection of surface defects in steel. In this paper, we propose an improved high-speed and high-precision Efficient Fusion Coordination network (EFC-YOLO) without increasing the model's size. Since modifications to enhance feature extraction in shallow networks tend to affect the speed of model inference, in order to simultaneously ensure the accuracy and speed of detection, we add the improved Fusion-Faster module to the backbone network of YOLOv7. Partial Convolution (PConv) serves as the basic operator of the module, which strengthens the feature-extraction ability of shallow networks while maintaining speed. Additionally, we incorporate the Shortcut Coordinate Attention (SCA) mechanism to better capture the location information dependency, considering both lightweight design and accuracy. The de-weighted Bi-directional Feature Pyramid Network (BiFPN) structure used in the neck part of the network improves the original Path Aggregation Network (PANet)-like structure by adding step branches and reducing computations, achieving better feature fusion. In the experiments conducted on the NEU-DET dataset, the final model achieved an 85.9% mAP and decreased the GFLOPs by 60%, effectively balancing the model's size with the accuracy and speed of detection. [ABSTRACT FROM AUTHOR]

With the popularity of ChatGPT, there has been increasing attention towards dialogue systems. Researchers are dedicated to designing a knowledgeable model that can engage in conversations like humans. Traditional seq2seq dialogue models often suffer from limited performance and the issue of generating safe responses. In recent years, large-scale pretrained language models have demonstrated their powerful capabilities across various domains. Many studies have leveraged these pretrained models for dialogue tasks to address concerns such as safe response generation. Pretrained models can enhance responses by carrying certain knowledge information after being pre-trained on large-scale data. However, when specific knowledge is required in a particular domain, the model may still generate bland or inappropriate responses, and the interpretability of such models is poor. Therefore, in this paper, we propose the KRP-DS model. We design a knowledge module that incorporates a knowledge graph as external knowledge in the dialogue system. The module utilizes contextual information for path reasoning and guides knowledge prediction. Finally, the predicted knowledge is used to enhance response generation. Experimental results show that our proposed model can effectively improve the quality and diversity of responses while having better interpretability, and outperforms baseline models in both automatic and human evaluations. [ABSTRACT FROM AUTHOR]

In the context of COVID-19, the research on various aspects of the venipuncture robot field has become increasingly hot, but there has been little research on robotic needle insertion angles, primarily performed at a rough angle. This will increase the rate of puncture failure. Furthermore, there is sometimes significant pain due to the patients' differences. This paper investigates the optimal needle entry angle decision for a dorsal hand intravenous injection robot. The dorsal plane of the hand was obtained by a linear structured light scan, which was used as a basis for calculating the needle entry angle. Simulation experiments were also designed to determine the optimal needle entry angle. Firstly, the linear structured optical system was calibrated and optimized, and the error function was constructed and solved iteratively by the optimization method to eliminate measurement error. Besides, the dorsal hand was scanned to obtain the spatial point clouds of the needle entry area, and the least squares method was used to fit it to obtain the dorsal hand plane. Then, the needle entry angle was calculated based on the needle entry area plane. Finally, the changes in the penetration force under different needle entry angles were analyzed to determine the optimal needle insertion angle. According to the experimental results, the average error of the optimized structured light plane position was about 0.1 mm, which meets the needs of the project, and a large angle should be properly selected for needle insertion during the intravenous injection. [ABSTRACT FROM AUTHOR]

Aspect-level sentiment classification has been widely used by researchers as a fine-grained sentiment classification task to predict the sentiment polarity of specific aspect words in a given sentence. Previous studies have shown relatively good experimental results using graph convolutional networks, so more and more approaches are beginning to exploit sentence structure information for this task. However, these methods do not link aspect word and context well. To address this problem, we propose a method that utilizes a hierarchical multi-head attention mechanism and a graph convolutional network (MHAGCN). It fully considers syntactic dependencies and combines semantic information to achieve interaction between aspect words and context. To fully validate the effectiveness of the method proposed in this paper, we conduct extensive experiments on three benchmark datasets, which, according to the experimental results, show that the method outperforms current methods. [ABSTRACT FROM AUTHOR]

Knowledge Graph (KG) is widely used for recommendation tasks due to its rich semantic information and external structure. Current knowledge graph recommendation models are insufficient for the learning of user–item interactive behaviors as well as external knowledge, and ignore the noise information in KG. To alleviate these limitations, we present a Collaborative Relation-aware Attention Network with Differentiable sampling (DCRAN) that can fully learn user–item interactions and knowledge graphs, and accurately select recommendation-related entities in KG. Specifically, DCRAN explicitly learns the embeddings of users and items and encodes them into collaborative interaction signals, which are used as guidance signals to extract knowledge. Furthermore, DCRAN emphasizes the importance of KG relations and constructs a relation-aware attention network to learn the representation of items in KG. Most importantly, DCRAN applies a differentiable sampling strategy on entities, which can reduce triplets that have a negative effect on recommendation. Experimental results on three public datasets manifest the effectiveness of DCRAN. [ABSTRACT FROM AUTHOR]

The successful deployment of WLAN for high speed data transmission and 3G cellular systems for wide coverage and global roaming has emerged to be a complementary platform for wireless data communications. But security in the 3G-WLAN interworking, especially the efficient authentication and valid key management, has been remaining a challenging issue. What's more, some emerging security challenges are neglected by 3GPP specifications as well as the previous studies. This paper first analyzes and evaluates the current contributions in this field, and then puts forward some design issues. Thereafter, by modifying the EAP-AKA keying framework we propose an improved authentication scheme which enables a WLAN user to efficiently access packet switch services through the 3G networks. What's more, through the new keying framework the user can efficiently realize the future re-authentications and handover authentications. The proposed authentication scheme, the corresponding re-authentications and handover authentications are simulated, and results indicate that our scheme can reduce authentication latency significantly. [ABSTRACT FROM AUTHOR]

Existing word embeddings learning algorithms only employ the contexts of words, but different text documents use words and their relevant parts of speech very differently. Based on the preceding assumption, in order to obtain appropriate word embeddings and further improve the effect of text classification, this paper studies in depth a representation of words combined with their parts of speech. First, using the parts of speech and context of words, a more expressive word embeddings can be obtained. Further, to improve the efficiency of look‐up tables, we construct a two‐dimensional table that is in the format to represent words in text documents. Finally, the two‐dimensional table and a Bayesian theorem are used for text classification. Experimental results show that our model has achieved more desirable results on standard data sets. And it has more preferable versatility and portability than alternative models. [ABSTRACT FROM AUTHOR]