Implicit discourse relation recognition (IDRR) aims to infer logical semantic relations between two adjacent text spans (also named arguments) without explicit indictive connectives, which is crucial to discourse analysis. Previous methods primarily focus on capturing the semantic features of discourse or the complex interaction patterns between the two arguments, including the utilization of pre-trained language models (PLMs). However, these approaches often overlook the fact that the semantic understanding of an argument cannot be interpreted independently from the overall paragraph-level discourse structure. Additionally, information about paragraph structure can help organize the context to establish semantic coherence in discourse from the perspective of cognitive linguistics, thereby creating a clear cognitive framework for the reader. Therefore, we propose a multi-hierarchy graph convolutional network framework based on paragraph-level discourse units for IDRR. Unlike conventional methods that utilize arguments as language units, we explore discourse units (DUs) within the broader context of a paragraph. Specifically, we employ the Stanford Parser to capture the syntactic dependency clues of each sentence within the DU and integrate them into the sentence representation learning. To further extract the structural information of paragraphs within the discourse, we leverage a multi-hierarchy graph convolutional network to derive the DU representation enriched with structural cues. Subsequently, we adopt multi-head attention to obtain important and structurally enhanced semantic features. The experimental results on the Penn Discourse TreeBank (PDTB) demonstrate that our model achieves performance comparable to that of benchmark models.

Semantic segmentation is crucial in autonomous driving, offering exceptional scene understanding to tackle challenges like small object edges and blurred textures in complex traffic environments. By performing pixel-level classification, it provides vehicles with comprehensive environmental information, ensuring safe navigation. However, when dealing with fine, fuzzy-bordered objects in complex scenes, the existing techniques face the issue of low segmentation accuracy due to their insufficient feature extraction capability. To address this problem, this study proposes a semantic segmentation network that combines Gaussian perception with iterative multi-scale attention. The method integrates Gaussian perception with local–global channel attention, accurately models pixel associations, and dynamically focuses on features to address the issue of edge blurring in complex scene segmentation. At the same time, the method employs the difference module to enhance the low-frequency features, and integrates the iterative multi-scale attention mechanism to achieve deep integration of low-frequency and high-frequency information. This enhances the fine capture of features and mitigates the edge discontinuity issue in segmentation caused by the masking of boundary information. In addition, the method combines channel and spatial attention to optimize feature extraction, enhance the sensory field, and improve detail, context, and boundary recognition abilities. This significantly improves the feature expression ability and reduces the probability of background mis-segmentation. The experimental results show that the proposed method achieves 79.34% mIoU on the Cityscapes validation set (a 1.32% improvement over PIDNet-S) and 81.48% mIoU on the CamVid test set (a 1.05% improvement over PIDNet-S). These results demonstrate significant improvements over existing state-of-the-art methods, confirming the effectiveness of this approach in semantic segmentation of complex urban scenes. The source code has been made publicly available on GitHub: https://github.com/wgsheng897/GMSANet.git

Word Sense Disambiguation (WSD) aims to determine the meaning of the target word according to the given context. Currently, a single representation enhanced by glosses from different dictionaries or languages is used to characterize each word sense. By analyzing the similarity between glosses of the same word sense, we find semantic biases among them, revealing that the glosses have their own descriptive perspectives. Therefore, the traditional approach of integrating all glosses by a single representation results in failing to present the unique semantics revealed by the individual glosses. In this paper, a quantum superposition state is employed to formalize the representations of multiple glosses of the same word sense to reveal their distributions. Furthermore, the quantum interference model is leveraged to calculate the probability that the target word belongs to this superposition state. The advantage is that the interference term can be regarded as a confidence level to guide word sense recognition. Finally, experiments are performed under standard WSD evaluation framework and the latest cross-lingual datasets, and the results verify the effectiveness of our model.

Data imbalance, also known as the long-tail distribution of data, is an important challenge for data-driven models. In the Word Sense Disambiguation (WSD) task, the long-tail phenomenon of word sense distribution is more common, making it difficult to effectively represent and identify Long-Tail Senses (LTSs). Therefore exploring representation methods that do not rely heavily on the training sample size is an important way to combat LTSs. Considering that many new states, namely superposition states, can be constructed from several known states in quantum mechanics, superposition states provide the possibility to obtain more accurate representations from inferior representations learned from a small sample size. Inspired by quantum superposition states, a representation method in Hilbert space is proposed to reduce the dependence on large sample sizes and thus combat LTSs. We theoretically prove the correctness of the method, and verify its effectiveness under the standard WSD evaluation framework and obtain state-of-the-art performance. Furthermore, we also test on the constructed LTS and the latest cross-lingual datasets, and achieve promising results.

Sarcasm, a linguistic technique employed to express emotions opposite to their literal meaning, has garnered significant attention from researchers due to the rise of social media. Detecting sarcasm in a multi-modal context has become a focal point in recent studies. However, existing research primarily relies on identifying inconsistencies between text semantics and image semantics, often lacking a deep understanding of images. Consequently, capturing inconsistencies between images and texts poses a challenge in many cases. In this paper, we propose the Entity-Relational Graph Convolutional Network (ERGCN) as a solution to detect sarcasm by examining the relationship between entities within images. Our approach involves extracting entities and text descriptions from each image, which provides valuable entity information. Subsequently, we employ external knowledge to construct a cross-modal graph for each text and image pair, emphasizing the presence of internal contradictory information. Finally, we utilize the graph convolutional network to identify inconsistent information across modalities and successfully detect sarcasm. Experimental results demonstrate that our model achieves state-of-the-art performance on a widely used multimodal Twitter dataset.

The long-tailed distribution, also called the heavy-tailed distribution, is common in nature. Since both words and their senses in natural language have long-tailed phenomenon in usage frequency, the Word Sense Disambiguation (WSD) task faces serious data imbalance. The existing learning strategies or data augmentation methods are difficult to deal with the lack of training samples caused by the single application scenario of long-tail senses, and the word sense representations caused by unique word sense definitions. Considering that the features extracted from the Disentangled Representation (DR) independently describe the essential properties of things, and DR does not require deep feature extraction and fusion processes, it alleviates the dependence of the representation learning on the training samples. We propose a novel DR by constraining the covariance matrix of a multivariate Gaussian distribution, which can enhance the strength of independence among features compared to β-VAE. The WSD model implemented by the reinforced DR outperforms the baselines on the English all-words WSD evaluation framework, the constructed long-tail word sense datasets, and the latest cross-lingual datasets.

Recognizing implicit discourse relation is a challenging task in discourse analysis, which aims to understand and infer the latent relations between two discourse arguments, such as temporal, comparison. Most of the present models largely focus on learning-based methods that utilize only intra-sentence textual information to identify discourse relations, ignoring the wider contexts beyond the discourse. Moreover, people comprehend the meanings and the relations of discourses, heavily relying on their interconnected working memories (e.g., instant memory, long-term memory). Inspired by this, we propose a Knowledge-Enhanced Attentive Neural Network (KANN) framework to address these issues. Specifically, it establishes a mutual attention matrix to capture the reciprocal information between two arguments, as instant memory. While implicitly stated knowledge in the arguments is retrieved from external knowledge source and encoded as inter-words semantic connection embeddings to further construct knowledge matrix, as long-term memory. We devise a novel paradigm with two ways by the collaboration of the memories to enrich the argument representation: 1) integrating the knowledge matrix into the mutual attention matrix, which implicitly maps knowledge into the process of capturing asymmetric interactions between two discourse arguments; 2) directly concatenating the argument representations and the semantic connection embeddings, which explicitly supplements knowledge to help discourse understanding. The experimental results on the PDTB also show that our KANN model is effective.

Implicit discourse relation recognition is a challenging task due to the lack of connectives as strong linguistic clues.Previous methods primarily encode two arguments separately or extract the specific interaction patterns for the task, which have not fully exploited the annotated relation signal.Therefore, we propose a novel TransS-driven joint learning architecture to address the issues.Specifically, based on the multi-level encoder, we 1) translate discourse relations in low-dimensional embedding space (called TransS), which could mine the latent geometric structure information of argumentrelation instances; 2) further exploit the semantic features of arguments to assist discourse understanding; 3) jointly learn 1) and 2) to mutually reinforce each other to obtain the better argument representations, so as to improve the performance of the task.Extensive experimental results on the Penn Discourse TreeBank (PDTB) show that our model achieves competitive results against several state-of-the-art systems.

Implicit discourse relation recognition aims to understand and annotate the latent relations between two discourse arguments, such as temporal, comparison, etc. Most previous methods encode two discourse arguments separately, the ones considering pair specific clues ignore the bidirectional interactions between two arguments and the sparsity of pair patterns. In this paper, we propose a novel neural Tensor network framework with Interactive Attention and Sparse Learning (TIASL) for implicit discourse relation recognition. (1) We mine the most correlated word pairs from two discourse arguments to model pair specific clues, and integrate them as interactive attention into argument representations produced by the bidirectional long short-term memory network. Meanwhile, (2) the neural tensor network with sparse constraint is proposed to explore the deeper and the more important pair patterns so as to fully recognize discourse relations. The experimental results on PDTB show that our proposed TIASL framework is effective.

This paper presents an automatic detection system capable of detecting an automobile dashboard with high accuracy. Since the structure of an automobile dashboard is quite different from general instruments, commonly used algorithms for instrument detection can hardly meet the accuracy and robustness. In this paper, a novel approach is presented to detect an automobile dashboard. The contour retrieving algorithm is first performed to extract contour image of dashboard. Progressive Probabilistic Hough Transform (PPHT) is then applied to fit accurate pointer position. A K-means clustering based method is then proposed recognizing tick marks. Since the result of K-means clustering always includes noise items, Z-test is introduced to insure precise clustering result. Lagrange interpolation method is used to describe the precise linear relationship between angle and reading of the instrument. The experimental test shows robustness and precision of this detection algorithm.