Understanding Tasks Research Articles

Multi-task learning approach for utilizing temporal relations in natural language understanding tasks

Various studies have been conducted on multi-task learning techniques in natural language understanding (NLU), which build a model capable of processing multiple tasks and providing generalized performance. Most documents written in natural languages contain time-related information. It is essential to recognize such information accurately and utilize it to understand the context and overall content of a document while performing NLU tasks. In this study, we propose a multi-task learning technique that includes a temporal relation extraction task in the training process of NLU tasks such that the trained model can utilize temporal context information from the input sentences. To utilize the characteristics of multi-task learning, an additional task that extracts temporal relations from given sentences was designed, and the multi-task model was configured to learn in combination with the existing NLU tasks on Korean and English datasets. Performance differences were analyzed by combining NLU tasks to extract temporal relations. The accuracy of the single task for temporal relation extraction is 57.8 and 45.1 for Korean and English, respectively, and improves up to 64.2 and 48.7 when combined with other NLU tasks. The experimental results confirm that extracting temporal relations can improve its performance when combined with other NLU tasks in multi-task learning, compared to dealing with it individually. Also, because of the differences in linguistic characteristics between Korean and English, there are different task combinations that positively affect extracting the temporal relations.

English training requirements and associated factors for non-native English-speaking nurses: A critical gap analysis based on the importance-performance method

Aims and objectivesTo investigate the English training requirements, priorities, and related factors of non-native English-speaking nurses. BackgroundFew studies have focused on the English training requirements of nurses in non-native English-speaking hospitals, and even fewer applied quantitative methods to analyze their English needs and related factors. DesignA total of 397 clinical nurses from a hospital in Zhejiang Province, China, were invited to answer questions from the 17-item English Language Requirement Scale (ELRS-17) through an online questionnaire system from May 7–12, 2021. MethodsThe importance-performance analysis (IPA) method was used to identify the critical training requirement gaps in the English skills of non-native English-speaking nurses in the case hospital. ResultsThe results of requirements showed that looking up foreign literature, writing medical/nursing academic articles and reports, and attending international medical/nursing academic conferences were the top three English learning purposes for nurses. Critical gap analysis with the IPA method revealed that medical dialogues (e.g., morning shift conversations), speeches (e.g., academic symposia), and everyday talk (e.g., telephoning and greetings) are very important yet inadequately trained skills for nurses at present, while nurses are adequately competent at the important tasks of understanding medical/nursing lectures and courses and oral international academic reports and reading academic articles and reports. ConclusionThe results of this study indicate that English training requirements for non-native English-speaking nurses revolve around facilitating contribution to nursing research and conference attendance, while more focus on spoken English is needed. Hospital decision makers can better understand the requirements and current performance of English language training for non-native English-speaking nurses. Furthermore, a suitable training plan and corresponding content can be designed for nurses.

Bridging the Gap between Synthetic and Natural Questions via Sentence Decomposition for Semantic Parsing

Abstract Semantic parsing maps natural language questions into logical forms, which can be executed against a knowledge base for answers. In real-world applications, the performance of a parser is often limited by the lack of training data. To facilitate zero-shot learning, data synthesis has been widely studied to automatically generate paired questions and logical forms. However, data synthesis methods can hardly cover the diverse structures in natural languages, leading to a large gap in sentence structure between synthetic and natural questions. In this paper, we propose a decomposition-based method to unify the sentence structures of questions, which benefits the generalization to natural questions. Experiments demonstrate that our method significantly improves the semantic parser trained on synthetic data (+7.9% on KQA and +8.9% on ComplexWebQuestions in terms of exact match accuracy). Extensive analysis demonstrates that our method can better generalize to natural questions with novel text expressions compared with baselines. Besides semantic parsing, our idea potentially benefits other semantic understanding tasks by mitigating the distracting structure features. To illustrate this, we extend our method to the task of sentence embedding learning, and observe substantial improvements on sentence retrieval (+13.1% for Hit@1).

Object-agnostic Affordance Categorization via Unsupervised Learning of Graph Embeddings

Acquiring knowledge about object interactions and affordances can facilitate scene understanding and human-robot collaboration tasks. As humans tend to use objects in many different ways depending on the scene and the objects’ availability, learning object affordances in everyday-life scenarios is a challenging task, particularly in the presence of an open set of interactions and objects. We address the problem of affordance categorization for class-agnostic objects with an open set of interactions; we achieve this by learning similarities between object interactions in an unsupervised way and thus inducing clusters of object affordances. A novel depth-informed qualitative spatial representation is proposed for the construction of Activity Graphs (AGs), which abstract from the continuous representation of spatio-temporal interactions in RGB-D videos. These AGs are clustered to obtain groups of objects with similar affordances. Our experiments in a real-world scenario demonstrate that our method learns to create object affordance clusters with a high V-measure even in cluttered scenes. The proposed approach handles object occlusions by capturing effectively possible interactions and without imposing any object or scene constraints.

GeoSynth: A Photorealistic Synthetic Indoor Dataset for Scene Understanding.

Deep learning has revolutionized many scene perception tasks over the past decade. Some of these improvements can be attributed to the development of large labeled datasets. The creation of such datasets can be an expensive, time-consuming, and imperfect process. To address these issues, we introduce GeoSynth, a diverse photorealistic synthetic dataset for indoor scene understanding tasks. Each GeoSynth exemplar contains rich labels including segmentation, geometry, camera parameters, surface material, lighting, and more. We demonstrate that supplementing real training data with GeoSynth can significantly improve network performance on perception tasks, like semantic segmentation. A subset of our dataset will be made publicly available at https://github.com/geomagical/GeoSynth.

Understanding Chinese Moral Stories with Further Pre-Training

The goal of moral understanding is to grasp the theoretical concepts embedded in a narrative by delving beyond the concrete occurrences and dynamic personas. Specifically, the narrative is compacted into a single statement without involving any characters within the original text, necessitating a more astute language model that can comprehend connotative morality and exhibit commonsense reasoning. The “pretraining + fine-tuning” paradigm is widely embraced in neural language models. In this paper, we propose an intermediary phase to establish an improved paradigm of “pre-training + further pre-training + finetuning”. Further pre-training generally refers to continual learning on task-specific or domain-relevant corpora before being applied to target tasks, which aims at bridging the gap in data distribution between the phases of pre-training and fine-tuning. Our work is based on a Chinese dataset named STORAL-ZH that composes of 4k human-written story-moral pairs. Furthermore, we design a two-step process of domain-adaptive pre-training in the intermediary phase. The first step depends on a newly-collected Chinese dataset of Confucian moral culture. And the second step bases on the Chinese version of a frequently-used commonsense knowledge graph (i.e. ATOMIC) to enrich the backbone model with inferential knowledge besides morality. By comparison with several advanced models including BERTbase, RoBERTa-base and T5-base, experimental results on two understanding tasks demonstrate the effectiveness of our proposed three-phase paradigm.

Understanding common human driving semantics for autonomous vehicles

Autonomous vehicles will share roads with human-driven vehicles until the transition to fully autonomous transport systems is complete. The critical challenge of improving mutual understanding between both vehicle types cannot be addressed only by feeding extensive driving data into data-driven models but by enabling autonomous vehicles to understand and apply common driving behaviors analogous to human drivers. Therefore, we designed and conducted two electroencephalography experiments for comparing the cerebral activities of human linguistics and driving understanding. The results showed that driving activates hierarchical neural functions in the auditory cortex, which is analogous to abstraction in linguistic understanding. Subsequently, we proposed a neural-informed, semantics-driven framework to understand common human driving behavior in a brain-inspired manner. This study highlights the pathway of fusing neuroscience into complex human behavior understanding tasks and provides a computational neural model to understand human driving behaviors, which will enable autonomous vehicles to perceive and think like human drivers.

Improving Gender-Related Fairness in Sentence Encoders: A Semantics-Based Approach

The ever-increasing number of systems based on semantic text analysis is making natural language understanding a fundamental task: embedding-based language models are used for a variety of applications, such as resume parsing or improving web search results. At the same time, despite their popularity and widespread use, concern is rapidly growing due to their display of social bias and lack of transparency. In particular, they exhibit a large amount of gender bias, favouring the consolidation of social stereotypes. Recently, sentence embeddings have been introduced as a novel and powerful technique to represent entire sentences as vectors. We propose a new metric to estimate gender bias in sentence embeddings, named bias score. Our solution leverages semantic importance of words and previous research on bias in word embeddings, and it is able to discern between neutral and biased gender information at sentence level. Experiments on a real-world dataset demonstrate that our novel metric can identify gender stereotyped sentences. Furthermore, we employ bias score to detect and then remove or compensate for the more stereotyped entries in text corpora used to train sentence encoders, improving their degree of fairness. Finally, we prove that models retrained on fairer corpora are less prone to make stereotypical associations compared to their original counterpart, while preserving accuracy in natural language understanding tasks. Additionally, we compare our experiments with traditional methods for reducing bias in embedding-based language models.

TME: Tree-guided Multi-task Embedding Learning towards Semantic Venue Annotation

The prevalence of location-based services has generated a deluge of check-ins, enabling the task of human mobility understanding. Among the various types of information associated with the check-in venues, categories (e.g., Bar and Museum ) are vital to the task, as they often serve as excellent semantic characterization of the venues. Despite its significance and importance, a large portion of venues in the check-in services do not have even a single category label, such as up to 30% of venues in the Foursquare system lacking category labels. We, therefore, address the problem of semantic venue annotation, i.e., labeling the venue with a semantic category. Existing methods either fail to fully exploit the contextual information in the check-in sequences, or do not consider the semantic correlations across related categories. As such, we devise a Tree-guided Multi-task Embedding model (TME for short) to learn effective representations of venues and categories for the semantic annotation. TME jointly learns a common feature space by modeling multi-contexts of check-ins and utilizes the predefined category hierarchy to regularize the relatedness among categories. We evaluate TME over the task of semantic venue annotation on two check-in datasets. Experimental results show the superiority of TME over several state-of-the-art baselines.

Quantitatively Mapping Discolored Seawater around Submarine Volcanoes Using Satellite GCOM-C SGLI Data: A Case Study of the Krakatau Eruption in Indonesia in December 2018

The final goal of this paper is to contribute to the difficult task of understanding and forecasting submarine volcanic eruption activity by proposing a method to quantify discolored water. To achieve this purpose, we quantitatively analyzed the discolored seawater seen before and after the eruption of the marine environment around the Indonesian submarine volcano “Anak Krakatau”, which erupted at the end of December 2018, from the viewpoint of the “dominant wavelength”. The atmospherically corrected COM-C SGLI data for 17 periods from the eruption from October 2018 to March 2019 were used. As a result, the following three main items were found. First, the average ± standard deviation of the entire dominant wavelength was 497 nm ± 2 nm before the eruption and 515 nm ± 35 nm after the eruption. Second, the discolored water area around the island derived from SGLI was detected from the contour line with dominant wavelengths of 500 nm and 560 nm. Third, the size of a dominant wavelength of 500 nm or more in the discolored water areas changed in a complicated manner within the range of almost 0 to 35 km2. The area of the dominant wavelength of 500 nm or more slightly increased just before the eruption. Finally, it was proven that the “dominant wavelength” from the SGLI proposed in this paper can be a very effective tool in understanding or predicting submarine volcanic activity.

Deep Learning for Scene Flow Estimation on Point Clouds: A Survey and Prospective Trends

AbstractAiming at obtaining structural information and 3D motion of dynamic scenes, scene flow estimation has been an interest of research in computer vision and computer graphics for a long time. It is also a fundamental task for various applications such as autonomous driving. Compared to previous methods that utilize image representations, many recent researches build upon the power of deep analysis and focus on point clouds representation to conduct 3D flow estimation. This paper comprehensively reviews the pioneering literature in scene flow estimation based on point clouds. Meanwhile, it delves into detail in learning paradigms and presents insightful comparisons between the state‐of‐the‐art methods using deep learning for scene flow estimation. Furthermore, this paper investigates various higher‐level scene understanding tasks, including object tracking, motion segmentation, etc. and concludes with an overview of foreseeable research trends for scene flow estimation.

Document-level relation extraction with multi-layer heterogeneous graph attention network

Document-level relation extraction is an essential task in natural language understanding and requires the ability to fully learn different representations of entities and multi-hop reasoning across multiple sentences. Most existing methods use graph neural networks to extract multi-hop relations. They only focus on the role of edge types and ignore node types when obtaining node representations. In addition, it is difficult for them to adequately capture the complex interactions in a document when converting a document into a graph. These problems will affect the performance of relation extraction. To solve the above problems, this paper proposes a document-level relation extraction model based on a multi-layer heterogeneous graph neural network (MHGNN). Different from previous methods that only focus on edge types between two nodes, a node type and edge type oriented heterogeneous graph attention network (NEHGAN) is proposed to further learn rich node representations. In addition, the model constructs three graphs, namely a word-level graph, a mention-level graph and an entity-level graph, for capturing the complex interactions between words, between mentions and between entities in the whole document. Experimental results demonstrate that MHGNN can effectively improve the performance of document-level relation extraction, and outperforms the state-of-the-art model by 1.04%, 1.03%, 1.69% and 0.6% in F1 on the development set and the test set of two public datasets, respectively.

The sensus fidei and Synodality: Theological Epistemology and the munus propheticum

The sensus fidei and Synodality: Theological Epistemology and the munus propheticum Bernhard Blankenhorn O.P. IN THE YEARS before and after the Second Vatican Council, Yves Congar famously called for a renewed ecclesiology that integrated both Christological and pneumatological approaches to the Church. The French Dominican sought a synthetic vision in which the themes of the Church’s apostolicity and the hierarchical powers transmitted by Christ would be framed within a rich understanding of the Church as a communion of grace, in which the Church’s perennity would not exclude her historicity or the Spirit’s work today.1 One can approach the overall program of advancing synodality as a way better to accentuate the pneumatological element of ecclesial life. Congar’s vast corpus of ecclesiological writings remains a treasure for the Church today, partly because of his astounding grasp of the theological tradition, and partly because his systematics remained animated by the properly theological task of faith seeking understanding. In this way, from his early writings on the theology of the laity forward, Congar exemplified a way of thinking about synodality and its pneumatological potential: to draw widely from the Church’s long [End Page 311] tradition, to acknowledge the historicity of the Church’s structures while refusing to give strictly sociological or political categories the last word, and to avoid one-sided theological solutions (such as a return to early modern Christo-monist ecclesiologies that reduce the laity to passive listeners of a teaching Church, or Pneuma-monist approaches that evacuate the apostolic tradition or ecclesial hierarchy altogether). The present article adopts a Congarian approach to believers’ supernatural sense of the faith (sensus fidei), specifically, by a modest proposal as to how we can best articulate the way in which this supernatural mode of cognition comes about. This article proposes the outline of a subchapter of theological epistemology in view of today’s ecclesial and theological debate on the true meaning of synodality. The theological payoff for seeking such an epistemology is manifold. Much confusion reigns around the question of how exactly the sensus fidei can be identified. Should we especially consult practicing Catholics, such as the laity most engaged in parochial life and evangelization, or faithful members of religious orders? What about fallen-away or barely practicing Catholics? What are we to make of voices on the margins, among so-called progressives or traditionalists? Do charismatic gifts inform the sensus fidei? And more broadly, how are we to interpret the sensus fidei in light of Scripture, Tradition, and magisterial teaching? The questions surrounding the role of the sensus fidei and its noetic character touch on numerous core ecclesiological issues. I. Preliminary Considerations The link between the sensus fidei and synodality has emerged with some frequency in recent magisterial documents and theological literature. Pope Francis’s apostolic constitution on the Synod of Bishops, Episcopalis Communio (2018) makes this connection explicit, and we find it several times in the International Theological Commission’s document Synodality in [End Page 312] the Life and Mission of the Church (2018).2 The same commission’s other recent text, Sensus fidei in the Life of the Church (2014) offers the most extensive teaching to date on the sensus fidei by a body of the Roman curia.3 Several recent collections of theological essays and monographs also take up this link, directly or indirectly.4 Still, much work remains to be done on the sensus fidei in general, and on its noetic aspects in particular.5 An epistemological lacuna remains in the theology of the sensus fidei, though part of its foundation has been laid. The work of John Henry Newman on the need to consult the lay faithful and Yves Congar’s preconciliar writings on the place of the laity in the Church helped to inspire the fathers of Vatican II to overcome the premodern dichotomy between the teaching and the listening Church.6 Newman’s theology on this question arose from his historical studies of the Arian crisis, while his [End Page 313] conviction about the importance of consulting the laity is linked to his theology of faith as a work of grace in man’s conscience and the...

Геномика человека: основные вехи развития с позиций правового подхода

The article describes stages of development of human genomics, which are the most significant ones for understanding of readiness of the society for consequences of the genetic revolution that has already started. Within the framework of such problem setting, the legal science faces the task of understanding whether the desire of the mankind to overcome genetic inequality of people will lead to even more inadmissible inequality forms.

Efficient Long-Text Understanding with Short-Text Models

AbstractTransformer-based pretrained language models (LMs) are ubiquitous across natural language understanding, but cannot be applied to long sequences such as stories, scientific articles, and long documents due to their quadratic complexity. While a myriad of efficient transformer variants have been proposed, they are typically based on custom implementations that require expensive pretraining from scratch. In this work, we propose SLED: SLiding-Encoder and Decoder, a simple approach for processing long sequences that re-uses and leverages battle-tested short-text pretrained LMs. Specifically, we partition the input into overlapping chunks, encode each with a short-text LM encoder and use the pretrained decoder to fuse information across chunks (fusion-in-decoder). We illustrate through controlled experiments that SLED offers a viable strategy for long text understanding and evaluate our approach on SCROLLS, a benchmark with seven datasets across a wide range of language understanding tasks. We find that SLED is competitive with specialized models that are up to 50x larger and require a dedicated and expensive pretraining step.

Continuous frame motion sensitive self-supervised collaborative network for video representation learning

Motion, as a feature of video that changes in temporal sequences, is crucial to visual understanding. The powerful video representation and extraction models are typically able to focus attention on motion features in challenging dynamic environments to complete more complex video understanding tasks. However, previous approaches discriminate mainly based on similar features in the spatial or temporal domain, ignoring the interdependence of consecutive video frames. In this paper, we propose the motion sensitive self-supervised collaborative network, a video representation learning framework that exploits a pretext task to assist feature comparison and strengthen the spatiotemporal discrimination power of the model. Specifically, we first propose the motion-aware module, which extracts consecutive motion features from the spatial regions by frame difference. The global–local contrastive module is then introduced, with context and enhanced video snippets being defined as appropriate positive samples for a broader feature similarity comparison. Finally, we introduce the snippet operation prediction module, which further assists contrastive learning to obtain more reliable global semantics by sensing changes in continuous frame features. Experimental results demonstrate that our work can effectively extract robust motion features and achieve competitive performance compared with other state-of-the-art self-supervised methods on downstream action recognition and video retrieval tasks.

Compositional Prompting Video-language Models to Understand Procedure in Instructional Videos

Instructional videos are very useful for completing complex daily tasks, which naturally contain abundant clip-narration pairs. Existing works for procedure understanding are keen on pretraining various video-language models with these pairs and then fine-tuning downstream classifiers and localizers in predetermined category space. These video-language models are proficient at representing short-term actions, basic objects, and their combinations, but they are still far from understanding long-term procedures. In addition, the predetermined procedure category faces the problem of combination disaster and is inherently inapt to unseen procedures. Therefore, we propose a novel compositional prompt learning (CPL) framework to understand long-term procedures by prompting short-term video-language models and reformulating several classical procedure understanding tasks into general video-text matching problems. Specifically, the proposed CPL consists of one visual prompt and three compositional textual prompts (including the action prompt, object prompt, and procedure prompt), which could compositionally distill knowledge from short-term video-language models to facilitate long-term procedure understanding. Besides, the task reformulation enables our CPL to perform well in all zero-shot, few-shot, and fully-supervised settings. Extensive experiments on two widely-used datasets for procedure understanding demonstrate the effectiveness of the proposed approach.

Cognitive flexibility moderates the association between theory of mind and children’s offer in the Ultimatum Game (La flexibilidad cognitiva modera la relación entre la teoría de la mente y las ofertas de los niños en el Juego del Ultimátum)

ABSTRACT This study examined the cognitive predictors and constraints of children’s strategic offers in the Ultimatum Game. The Ultimatum Game is a one-shot negotiation on the division of a given sum of rewards between two people, where the receiver has the right to punish the proposer by rejecting unfair offers, which results in nil reward for either party. Ninety-four Hong Kong children ranged between three and six years of age (48 girls, M = 4.94, SD = 0.90) participated in an Ultimatum Game, and tasks measuring theory of mind, cognitive flexibility, inhibitory control and verbal ability. Children’s theory of mind, measured with false belief understanding tasks, was positively associated with their Ultimatum Game offers. Regression analysis further demonstrated that cognitive flexibility moderated the association between false belief understanding and children’s offers in the Ultimatum Game. Being able to take the opponent’s perspective prompted children to offer more in the Ultimatum Game, but only when their cognitive flexibility was low.

Evaluating Deep Learning Techniques for Natural Language Inference

Natural language inference (NLI) is one of the most important natural language understanding (NLU) tasks. NLI expresses the ability to infer information during spoken or written communication. The NLI task concerns the determination of the entailment relation of a pair of sentences, called the premise and hypothesis. If the premise entails the hypothesis, the pair is labeled as an “entailment”. If the hypothesis contradicts the premise, the pair is labeled a “contradiction”, and if there is not enough information to infer a relationship, the pair is labeled as “neutral”. In this paper, we present experimentation results of using modern deep learning (DL) models, such as the pre-trained transformer BERT, as well as additional models that relay on LSTM networks, for the NLI task. We compare five DL models (and variations of them) on eight widely used NLI datasets. We trained and fine-tuned the hyperparameters for each model to achieve the best performance for each dataset, where we achieved some state-of-the-art results. Next, we examined the inference ability of the models on the BreakingNLI dataset, which evaluates the model’s ability to recognize lexical inferences. Finally, we tested the generalization power of our models across all the NLI datasets. The results of the study are quite interesting. In the first part of our experimentation, the results indicate the performance advantage of the pre-trained transformers BERT, RoBERTa, and ALBERT over other deep learning models. This became more evident when they were tested on the BreakingNLI dataset. We also see a pattern of improved performance when the larger models are used. However, ALBERT, given that it has 18 times fewer parameters, achieved quite remarkable performance.

Improved Text Matching Model Based on BERT

Text matching is a basic and important task in natural language understanding, this paper proposes a new model BBMC for the problem of insufficient feature extraction ability of existing text matching models, which integrates BiLSTM and multi-scale CNN on the basis of BERT. First, the word embedding representation of the text is obtained by the BERT, and then the semantic features of the text are further extracted by the double-layer BiLSTM, followed by the multi-scale CNN model, the key local features are extracted, and finally the linear and SoftMax function are used to classify. Experimental results on the LCQMC dataset show that the BBMC has been improved to a certain extent compared with other methods, and the accuracy on the test set can be best achieved 88.01%.