Semantic Parsing Research Articles

Dynamic multi teacher knowledge distillation for semantic parsing in KBQA

Knowledge base question answering (KBQA) is an important task of extracting answers from a knowledge base by analyzing natural language questions. Semantic parsing methods convert natural language questions into executable logical forms to obtain answers on the knowledge base. Conventional approaches often prioritize singular logical forms, overlooking the distinct strengths inherent in various logical frameworks for problem solving. Recognizing that different logical forms may excel in addressing specific types of questions, our aim is to harness these strengths. By integrating the strengths of different logical forms, we expect to achieve more comprehensive and effective semantic parsing solutions. In our paper, we propose a Dynamic Multi Teacher Knowledge Distillation for Semantic Parsing (DMTKD-SP). DMTKD-SP leverages a collection of teacher models, each mastering a unique logical form, to collaboratively guide a student model so that knowledge from different logical forms can be transferred into the student model. To achieve this, we employ a confidence-based weight assignment module to dynamically assign weights for each teacher model. Furthermore, we introduce a self-distillation mechanism to mitigate the confusion caused by simultaneous learning from multiple teachers. We evaluate DMTKD-SP across variants of the KQA Pro dataset, demonstrating an accuracy improvement of 0.35% on five types of questions, with a notable 0.75% improvement for Count questions.

Review of Convolutional Neural Network

Over the past decade, the domain of deep learning has emerged as a swiftly expanding area of interest among researchers globally. It offers substantial benefits over traditional shallow networks, particularly in the realms of feature extraction and model fitting. Deep learning excels at uncovering intricate, distributed features from raw data, which exhibit robust generalization capabilities. It has triumphantly addressed challenges that were once deemed intractable within the field of artificial intelligence. With the exponential growth in the volume of training data and a marked increase in computational power, deep learning has achieved remarkable strides and found applications across a spectrum of domains, including but not limited to object detection, computer vision, natural language processing, speech recognition, and semantic parsing, thereby accelerating the evolution of artificial intelligence. Deep learning encompasses a series of non-linear transformations organized hierarchically, with deep neural networks representing the predominant form of contemporary deep learning methodologies. Inspired by the neural connections found in the animal visual cortex, these networks are characterized by local connectivity, shared weights, and pooling operations. Such attributes not only simplify the network model and curtail the number of trainable parameters but also engender invariance to shifts, distortions, and scaling, thereby endowing the model with enhanced robustness and fault tolerance. This makes the training and optimization of the network structure more manageable. This paper commences with a retrospective on the evolution of convolutional neural networks (CNNs). Subsequently, it delineates the architectures of neuron models and multilayer perceptron. It proceeds to dissect the CNN architecture, which typically encompasses a sequence of convolutional and pooling layers, culminating in fully connected layers, each serving distinct functions. The paper also elaborates on various enhanced algorithms for CNNs, such as the “Network in Network” and “Spatial Transformer Networks,” while also introducing supervised and unsupervised learning methodologies pertinent to CNNs and highlighting several prevalent open-source tools. Further, the paper scrutinizes the application of CNNs in various fields, including image classification, facial recognition, audio retrieval, electrocardiogram analysis, and object detection. It posits the amalgamation of CNNs with recurrent neural networks as a potential alternative for training datasets. The research culminates in the design of CNN structures with varying parameters and depths, supported by experimental analysis that elucidates the interplay among these parameters and the ramifications of their configuration. Ultimately, the paper synthesizes the merits and lingering challenges associated with CNNs and their practical applications.

A Semantic Segmentation Method for Winter Wheat in North China Based on Improved HRNet

Winter wheat is one of the major crops for global food security. Accurate statistics of its planting area play a crucial role in agricultural policy formulation and resource management. However, the existing semantic segmentation methods for remote sensing images are subjected to limitations in dealing with noise, ambiguity, and intra-class heterogeneity, posing a negative impact on the segmentation performance of the spatial distribution and area of winter wheat fields in practical applications. In response to the above challenges, we proposed an improved HRNet-based semantic segmentation model in this paper. First, this model incorporates a semantic domain module (SDM), which improves the model’s precision of pixel-level semantic parsing and reduces the interference from noise through multi-confidence scale class representation. Second, a nested attention module (NAM) is embedded, which enhances the model’s capability of recognizing correct correlations in pixel classes. The experimental results show that the proposed model achieved a mean intersection over union (mIoU) of 80.51%, a precision of 88.64%, a recall of 89.14%, an overall accuracy (OA) of 90.12%, and an F1-score of 88.89% on the testing set. Compared to traditional methods, our model demonstrated better segmentation performance in winter wheat semantic segmentation tasks. The achievements of this study not only provide an effective tool and technical support for accurately measuring the area of winter wheat fields, but also have important practical value and profound strategic significance for optimizing agricultural resource allocation and achieving precision agriculture.

KF-VTON: Keypoints-Driven Flow Based Virtual Try-On Network

Image-based virtual try-on aims to fit a target garment to a reference person. Most existing methods are limited to solving the Garment-To-Person (G2P) try-on task that transfers a garment from a clean product image to the reference person and do not consider the Person-To-Person (P2P) try-on task that transfers a garment from a clothed person image to the reference person, which limits the practical applicability. The P2P try-on task is more challenging due to spatial discrepancies caused by different poses, body shapes, and views between the reference person and the target person. To address this issue, we propose a novel Keypoints-Driven Flow Based Virtual Try-On Network (KF-VTON) for handling both the G2P and P2P try-on tasks. Our KF-VTON has two key innovations: (1) We propose a new keypoints-driven flow based deformation model to warp the garment. This model establishes spatial correspondences between the target garment and reference person by combining the robustness of Thin-plate Spline (TPS) based deformation and the flexibility of appearance flow based deformation. (2) We investigate a powerful Context-aware Spatially Adaptive Normalization (CSAN) generative module to synthesize the final try-on image. Particularly, CSAN integrates rich contextual information with semantic parsing guidance to properly infer unobserved garment appearances. Extensive experiments demonstrate that our KF-VTON is capable of producing photo-realistic and high-fidelity try-on results for the G2P as well as P2P try-on tasks and surpasses previous state-of-the-art methods both quantitatively and qualitatively. Our code is available at https://github.com/OIUIU/KF-VTON .

Compositionality and Sentence Meaning: Comparing Semantic Parsing and Transformers on a Challenging Sentence Similarity Dataset

Abstract One of the major outstanding questions in computational semantics is how humans integrate the meaning of individual words into a sentence in a way that enables understanding of complex and novel combinations of words, a phenomenon known as compositionality. Many approaches to modelling the process of compositionality can be classified as either ‘vector-based’ models, in which the meaning of a sentence is represented as a vector of numbers, or ‘syntax-based’ models, in which the meaning of a sentence is represented as a structured tree of labelled components. A major barrier in assessing and comparing these contrasting approaches is the lack of large, relevant datasets for model comparison. This paper aims to address this gap by introducing a new dataset, STS3k, which consists of 2,800 pairs of sentences rated for semantic similarity by human participants. The sentence pairs have been selected to systematically vary different combinations of words, providing a rigorous test and enabling a clearer picture of the comparative strengths and weaknesses of vector-based and syntax-based methods. Our results show that when tested on the new STS3k dataset, state-of-the-art transformers poorly capture the pattern of human semantic similarity judgments, while even simple methods for combining syntax- and vector-based components into a novel hybrid model yield substantial improvements. We further show that this improvement is due to the ability of the hybrid model to replicate human sensitivity to specific changes in sentence structure. Our findings provide evidence for the value of integrating multiple methods to better reflect the way in which humans mentally represent compositional meaning.

Automatic Frame Analysis for the Evaluation of Text Influence

The research paper focuses on the mechanism of semantic parser for automatic text analysis and represents some results of analyzing news with a total volume of 2.2 million clauses. This mechanism models the processes of text comprehension and in this paper it is applied to the evaluation of the text influence on the reader. The parser constructs a semantic representation of the text and maps the meaning of each clause to a particular frame belonging either to the set of 300 emotional or to the set of 4500 rational frames. Comparing the frames of pro-state (federal) and blocked sources in Russia, the authors show that the most frequent frames in federal news represent an interactive dialog between the media and the speaker (note, add, recall, etc.), while the most frequent frames in blocked news introduce the speaker’s utterance that has become known to the news agencies (report, write, say, etc.). This reflects the remoteness of the editors of blocked news from Russian speakers. The federal media more often report about funny or unexpected purchases, as well as about how federal employees rescue people, and this agenda should make a positive impression. The blocked media, on the other hand, are more likely to mention court sentencing and death, which should have a negative impact. The natural-language inference constructed via frame combinations is shown to enable the recognition of emotional situations in derived meanings that are not explicitly expressed in the text and, thus, it allows to analyze automatically more compound mechanisms of manipulation.

Face shape transfer via semantic warping

Face reshaping aims to adjust the shape of a face in a portrait image to make the face aesthetically beautiful, which has many potential applications. Existing methods 1) operate on the pre-defined facial landmarks, leading to artifacts and distortions due to the limited number of landmarks, 2) synthesize new faces based on segmentation masks or sketches, causing generated faces to look dissatisfied due to the losses of skin details and difficulties in dealing with hair and background blurring, and 3) project the positions of the deformed feature points from the 3D face model to the 2D image, making the results unrealistic because of the misalignment between feature points. In this paper, we propose a novel method named face shape transfer (FST) via semantic warping, which can transfer both the overall face and individual components (e.g., eyes, nose, and mouth) of a reference image to the source image. To achieve controllability at the component level, we introduce five encoding networks, which are designed to learn feature embedding specific to different face components. To effectively exploit the features obtained from semantic parsing maps at different scales, we employ a straightforward method of directly connecting all layers within the global dense network. This direct connection facilitates maximum information flow between layers, efficiently utilizing diverse scale semantic parsing information. To avoid deformation artifacts, we introduce a spatial transformer network, allowing the network to handle different types of semantic warping effectively. To facilitate extensive evaluation, we construct a large-scale high-resolution face dataset, which contains 14,000 images with a resolution of 1024 × 1024. Superior performance of our method is demonstrated by qualitative and quantitative experiments on the benchmark dataset.

Interlingual Syntactic Parsing: An Optimized Head-Driven Parsing for English to Indian Language Machine Translation

In the era of Artificial Intelligence (AI), significant progress has been made by enabling machines to understand and communicate in human languages. Central to this progress are parsers, which play a vital role in syntactic analysis and support various Natural language Processing (NLP) applications, including Machine Translation and sentiment analysis. This paper introduces a robust implementation of an optimized Head-Driven Parser designed to advance NLP capabilities beyond the limitations of traditional Lexicalized Tree Adjoining Grammar (L-TAG) based Parser. Traditional parser, while effective, often struggle with the capturing complexities of natural languages, especially translation between English to Indian languages. By leveraging Bi-directional approach and Head-Driven techniques, this research offers a revolutionary enhancement in parsing frameworks. This method not only improves performance in syntactic analysis but also facilitates complex tasks such as discourse analysis and semantic parsing. This research involves experimentation the Bi-Directional Parser on a dataset of 15,000 sentences, resulting a reduction in derivation variations compared to conventional TAG Parsers. This advancement highlights how Head-Driven Parsing can overcome traditional constraints and provide more reliable linguistic analysis. The paper demonstrates how this new implementation not only builds on the strengths of L-TAG but also addresses its limitations and contributes to expanding the scope of Tree Adjoining Grammarbased methodologies and advancing the field of Machine Translation.

Building a Turkish UCCA dataset

Abstract Semantic representation is the task of conveying the meaning of a natural language utterance by converting it to a logical form that can be processed and understood by machines. It is utilised by many applications in natural language processing (NLP), particularly in tasks relevant to natural language understanding (NLU). Due to the widespread use of semantic parsing in NLP, many semantic representation schemes with different forms have been proposed; Universal Conceptual Cognitive Annotation (UCCA) is one of them. UCCA is a cross-lingual semantic annotation framework that allows easy annotation without requiring substantial linguistic knowledge. UCCA-annotated datasets have been released so far for English, French, German, Russian, and Hebrew. In this paper, we present a UCCA-annotated Turkish dataset of 400 sentences that are obtained from the METU-Sabanci Turkish Treebank. We provide the UCCA annotation specifications defined for the Turkish language so that it can be extended further. We followed a semi-automatic annotation approach, where an external semantic parser is utilised for the initial annotation of the dataset, which is manually revised by two annotators. We used the same semantic parser model to evaluate the dataset with zero-shot and few-shot learning, demonstrating that even a small sample set from the target language in the training data has a notable impact on the performance of the parser (15.6% and 2.5% gain over zero-shot for labelled and unlabelled results, respectively).

Application of large language models based on knowledge graphs in question-answering systems: A review

The integration of Knowledge Graphs (KGs) and Large Language Models (LLMs) is emerging as a transformative advancement in AI, particularly within Question Answering (QA) systems. Traditional QA systems, constrained by static knowledge bases, have struggled with multimodal queries and personalized responses. The deep integration of KGs and LLMs offers a novel approach, combining the structured, contextual understanding of KGs with the semantic parsing capabilities of LLMs. This review explores the methodologies, algorithms, datasets, and applications of KG-LLM integration in QA systems, highlighting key research that enhances model performance, reasoning, and domain-specific applications. It addresses technical challenges such as data consistency, semantic understanding, and system optimization. Future directions focus on improving intrinsic model performance and achieving deeper integration of KGs and LLMs. This comprehensive overview underscores the potential of KG-LLM synergy to significantly improve service quality across diverse industries, including education and healthcare.

Shift-Reduce Task-Oriented Semantic Parsing with Stack-Transformers

Intelligent voice assistants, such as Apple Siri and Amazon Alexa, are widely used nowadays. These task-oriented dialogue systems require a semantic parsing module in order to process user utterances and understand the action to be performed. This semantic parsing component was initially implemented by rule-based or statistical slot-filling approaches for processing simple queries; however, the appearance of more complex utterances demanded the application of shift-reduce parsers or sequence-to-sequence models. Although shift-reduce approaches were initially considered the most promising option, the emergence of sequence-to-sequence neural systems has propelled them to the forefront as the highest-performing method for this particular task. In this article, we advance the research on shift-reduce semantic parsing for task-oriented dialogue. We implement novel shift-reduce parsers that rely on Stack-Transformers. This framework allows to adequately model transition systems on the transformer neural architecture, notably boosting shift-reduce parsing performance. Furthermore, our approach goes beyond the conventional top-down algorithm: we incorporate alternative bottom-up and in-order transition systems derived from constituency parsing into the realm of task-oriented parsing. We extensively test our approach on multiple domains from the Facebook TOP benchmark, improving over existing shift-reduce parsers and state-of-the-art sequence-to-sequence models in both high-resource and low-resource settings. We also empirically prove that the in-order algorithm substantially outperforms the commonly used top-down strategy. Through the creation of innovative transition systems and harnessing the capabilities of a robust neural architecture, our study showcases the superiority of shift-reduce parsers over leading sequence-to-sequence methods on the main benchmark.

NLP-based smart decision making for business and academics

Natural Language Processing (NLP) systems enable machines to understand, interpret, and generate human-like language, bridging the gap between human communication and computer understanding. Natural Language Interface to Databases (NLIDB) and Natural Language Interface to Visualization (NLIV) systems are designed to enable non-technical users to retrieve and visualize data through natural language queries. However, these systems often face challenges in handling complex correlation and analytical questions, limiting their effectiveness for comprehensive data analysis. Additionally, current Business Intelligence (BI) tools also struggle with understanding the context and semantics of complex questions, further hindering their usability for strategic decision-making. Also, when building these models for generating the queries from natural language, the system handles only the semantic parsing issues as each column header is being changed manually to their normal names by all existing models which is time-consuming, tedious, and subjective.Recent studies reflect the need for attention to context, semantics, and especially ambiguities in dealing with natural language questions. To address this problem, the proposed architecture focuses on understanding the context, correlation-based semantic analysis, and removal of ambiguities using a novel approach. An Enhanced Longest Common Subsequence (ELCS) is suggested where existing LCS is modified with a memorization component for mapping the natural language question tokens with ambiguous table column headers. This can speed up the overall process as human intervention is not required to manually change the column headers. The same is evidenced by carrying out thorough experimentation and comparative study in terms of precision, recall, and F1 score. By synthesizing the latest advancements and addressing challenges, this paper has proved how NLP can significantly enhance the accuracy and efficiency of information retrieval and visualization, broadening the inclusivity and usability of NLIDB, NLIV, and BI systems.

Text summarization based on semantic graphs: an abstract meaning representation graph-to-text deep learning approach

Nowadays, due to the constantly growing amount of textual information, automatic text summarization constitutes an important research area in natural language processing. In this work, we present a novel framework that combines semantic graph representations along with deep learning predictions to generate abstractive summaries of single documents, in an effort to utilize a semantic representation of the unstructured textual content in a machine-readable, structured, and concise manner. The overall framework is based on a well defined methodology for performing semantic graph parsing, graph construction, graph transformations for machine learning models and deep learning predictions. The employed semantic graph representation focuses on using the model of abstract meaning representation. Several combinations of graph construction and graph transformation methods are investigated to specify the most efficient of them for the machine learning models. Additionally, a range of deep learning architectures is examined, including a sequence-to-sequence attentive network, reinforcement learning, transformer-based architectures, and pre-trained neural language models. In this direction, a semantic graph representation of an original text is extracted, and then the present framework formulates the problem as a graph-to-summary learning problem to predict a summary of an original text. To the best of our knowledge, this formulation of graph-to-summary predictions in abstractive text summarization, without other intermediate steps in the machine learning phase, has not been presented in the relevant literature. Another important contribution is an introduction of a measure for assessing the factual consistency of the generated summaries in an effort to provide a qualitative evaluation. To assess the framework, an extensive experimental procedure is presented that uses popular datasets to evaluate key aspects of the proposed approach. The obtained results exhibit promising performance, validating the robustness of the proposed framework.

SpaceLight: A Framework for Enhanced On-Orbit Navigation Imagery

In the domain of space rendezvous and docking, visual navigation plays a crucial role. However, practical applications frequently encounter issues with poor image quality. Factors such as lighting uncertainties, spacecraft motion, uneven illumination, and excessively dark environments collectively pose significant challenges, rendering recognition and measurement tasks during visual navigation nearly infeasible. The existing image enhancement methods, while visually appealing, compromise the authenticity of the original images. In the specific context of visual navigation, space image enhancement’s primary aim is the faithful restoration of the spacecraft’s mechanical structure with high-quality outcomes. To address these issues, our study introduces, for the first time, a dedicated unsupervised framework named SpaceLight for enhancing on-orbit navigation images. The framework integrates a spacecraft semantic parsing network, utilizing it to generate attention maps that pinpoint structural elements of spacecraft in poorly illuminated regions for subsequent enhancement. To more effectively recover fine structural details within these dark areas, we propose the definition of a global structure loss and the incorporation of a pre-enhancement module. The proposed SpaceLight framework adeptly restores structural details in extremely dark areas while distinguishing spacecraft structures from the deep-space background, demonstrating practical viability when applied to visual navigation. This paper is grounded in space on-orbit servicing engineering projects, aiming to address visual navigation practical issues. It pioneers the utilization of authentic on-orbit navigation images in the research, resulting in highly promising and unprecedented outcomes. Comprehensive experiments demonstrate SpaceLight’s superiority over state-of-the-art low-light enhancement algorithms, facilitating enhanced on-orbit navigation image quality. This advancement offers robust support for subsequent visual navigation.

SPReCHD: Four-Chamber Semantic Parsing Network for Recognizing Fetal Congenital Heart Disease in Medical Metaverse.

Echocardiography is essential for evaluating cardiac anatomy and function during early recognition and screening for congenital heart disease (CHD), a widespread and complex congenital malformation. However, fetal CHD recognition still faces many difficulties due to instinctive fetal movements, artifacts in ultrasound images, and distinctive fetal cardiac structures. These factors hinder capturing robust and discriminative representations from ultrasound images, resulting in CHD's low prenatal detection rate. Hence, we propose a multi-scale gated axial-transformer network (MSGATNet) to capture fetal four-chamber semantic information. Then, we propose a SPReCHD: four-chamber semantic parsing network for recognizing fetal CHD in the clinical treatment of the medical metaverse, integrating MSGATNet to segment and locate four-chamber arbitrary contours, further capturing distinguished representations for the fetal heart. Comprehensive experiments indicate that our SPReCHD is sufficient in recognizing fetal CHD, achieving a precision of 95.92%, a recall of 94%, an accuracy of 95%, and a F1 score of 94.95% on the test set, dramatically improving the fetal CHD's prenatal detection rate.

Research on domain ontology construction based on the content features of online rumors

Online rumors are widespread and difficult to identify, which bring serious harm to society and individuals. To effectively detect and govern online rumors, it is necessary to conduct in-depth semantic analysis and understand the content features of rumors. This paper proposes a TFI domain ontology construction method, which aims to achieve semantic parsing and reasoning of the rumor text content. This paper starts from the term layer, the frame layer, and the instance layer, and based on the reuse of the top-level ontology, the extraction of core literature content features, and the discovery of new concepts in the real corpus, obtains the core classes (five parent classes and 88 subclasses) of the rumor domain ontology and defines their concept hierarchy. Object properties and data properties are designed to describe relationships between entities or their features, and the instance layer is created according to the real rumor datasets. OWL language is used to encode the ontology, Protégé is used to visualize it, and SWRL rules and pellet reasoner are used to mine and verify implicit knowledge of the ontology, and judge the category of rumor text. This paper constructs a rumor domain ontology with high consistency and reliability.

Implementation of Text to Image using Diffusion Model

Text-to-image generation is a transformative field in artificial intelligence, aiming to bridge the semantic gap between textual descriptions and visual representations. This presents a comprehensive approach to tackle this challenging task. Leveraging the advancements in deep learning, natural language processing (NLP), and computer vision, this proposes a cutting-edge model for generating high-fidelity images from textual prompts. Trained on a vast and varied dataset of written descriptions and related images, this model combines an image decoder and a text encoder within a hierarchical framework. To enhance realism, this incorporates attention mechanisms and fine-grained semantic parsing. The model's performance is rigorously evaluated through both quantitative metrics and qualitative human assessments. Results demonstrate its ability to produce visually compelling and contextually accurate images across various domains, from natural scenes to specific object synthesis. This further explores applications in creative content generation, design automation, and virtual environments, showcasing the potential impact of our approach. Additionally, this releases a user-friendly API, empowering developers and designers to integrate our model into their projects, and fostering innovation and creativity. Key Words: image generation model, Deep learning, Natural language processing.

Cross-linguistically consistent semantic and syntactic annotation of child-directed speech

AbstractCorpora of child speech and child-directed speech (CDS) have enabled major contributions to the study of child language acquisition, yet semantic annotation for such corpora is still scarce and lacks a uniform standard. Semantic annotation of CDS is particularly important for understanding the nature of the input children receive and developing computational models of child language acquisition. For example, under the assumption that children are able to infer meaning representations for (at least some of) the utterances they hear, the acquisition task is to learn a grammar that can map novel adult utterances onto their corresponding meaning representations, in the face of noise and distraction by other contextually possible meanings. To study this problem and to develop computational models of it, we need corpora that provide both adult utterances and their meaning representations, ideally using annotation that is consistent across a range of languages in order to facilitate cross-linguistic comparative studies. This paper proposes a methodology for constructing such corpora of CDS paired with sentential logical forms, and uses this method to create two such corpora, in English and Hebrew. The approach enforces a cross-linguistically consistent representation, building on recent advances in dependency representation and semantic parsing. Specifically, the approach involves two steps. First, we annotate the corpora using the Universal Dependencies (UD) scheme for syntactic annotation, which has been developed to apply consistently to a wide variety of domains and typologically diverse languages. Next, we further annotate these data by applying an automatic method for transducing sentential logical forms (LFs) from UD structures. The UD and LF representations have complementary strengths: UD structures are language-neutral and support consistent and reliable annotation by multiple annotators, whereas LFs are neutral as to their syntactic derivation and transparently encode semantic relations. Using this approach, we provide syntactic and semantic annotation for two corpora from CHILDES: Brown’s Adam corpus (English; we annotate $$\approx$$ ≈ 80% of its child-directed utterances), all child-directed utterances from Berman’s Hagar corpus (Hebrew). We verify the quality of the UD annotation using an inter-annotator agreement study, and manually evaluate the transduced meaning representations. We then demonstrate the utility of the compiled corpora through (1) a longitudinal corpus study of the prevalence of different syntactic and semantic phenomena in the CDS, and (2) applying an existing computational model of language acquisition to the two corpora and briefly comparing the results across languages.

Incorporating knowledge for joint Chinese word segmentation and part-of-speech tagging with SynSemGCN

PurposeRecent trends have shown the integration of Chinese word segmentation (CWS) and part-of-speech (POS) tagging to enhance syntactic and semantic parsing. However, the potential utility of hierarchical and structural information in these tasks remains underexplored. This study aims to leverage multiple external knowledge sources (e.g. syntactic and semantic features, lexicons) through various modules for the joint task.Design/methodology/approachWe introduce a novel learning framework for the joint CWS and POS tagging task, utilizing graph convolutional networks (GCNs) to encode syntactic structure and semantic features. The framework also incorporates a pre-defined lexicon through a lexicon attention module. We evaluate our model on a range of public corpora, including CTB5, PKU and UD, the novel ZX dataset and the comprehensive CTB9 dataset.FindingsExperimental results on these benchmark corpora demonstrate the effectiveness of our model in improving the performance of the joint task. Notably, we find that syntax information significantly enhances performance, while lexicon information helps mitigate the issue of out-of-vocabulary (OOV) words.Originality/valueThis study introduces a comprehensive approach to the joint CWS and POS tagging task by combining multiple features. Moreover, the proposed framework offers potential adaptability to other sequence labeling tasks, such as named entity recognition (NER).

On compositional generalization of transformer-based neural machine translation

Neural networks have been shown to have deficiencies in the ability of compositional generalization while existing work has generally targeted semantic parsing tasks. In this paper, we study the challenge of machine translation and introduce a dedicated Chinese→English dataset, CoGniTion (Compositional Generalization Machine Translation). The training, validation, and test sets have 216k, 10k, and 10k sentence pairs, respectively, which are simple but contain a variety of phrases. More importantly, we construct 2160 novel compounds(phrases) unseen in the training data and each compound is embedded into 5 distinct contexts, forming the CG-test set with 10,800 samples. We evaluate the performance of Transformer models based on conventional metrics like BLEU and the proposed compound translation error rates, demonstrating their difficulty in handling compositional generalization. For the Transformer trained from scratch, 27.42% of the test sentences are translated incorrectly, and 62.88% of the compounds fail to be translated consistently. The pretrained language models (PLM) including the large language models (LLM) still exhibit approximately 30% inconsistency. Moreover, we conduct an empirical analysis of various factors that impact compound translation performance, and show that the performance of PLMs may be overestimated due to unavoidable composition exposure in pertaining data.