Language Question Research Articles

Visual Chain-of-Thought Prompting for Knowledge-Based Visual Reasoning

Knowledge-based visual reasoning remains a daunting task since it not only requires machines to interpret the concepts and relationships from visual scenes but also associate them with external world knowledge to conduct a chain of reasoning on open-world questions. Previous works, however, treat visual perception and language-based reasoning as two independent modules, failing to attend to both modules throughout all stages of reasoning. To this end, we propose Visual Chain-of-thought Prompting (VCTP) for knowledge-based reasoning, which involves the interaction between visual content and natural language in an iterative step-by-step reasoning manner. VCTP contains three stages, see, think, and confirm. The see stage scans the image and grounds the visual concept candidates with a visual perception model. The think stage adopts a pre-trained large language model (LLM) to attend to key visual concepts from natural language questions adaptively. It then transforms key visual context into text context for prompting with a visual captioning model, and adopts the LLM to generate the answer. The confirm stage further uses the LLM to generate the supporting rationale to the answer, which is then passed through a cross-modality classifier to verify that it’s consistent with the visual context. We iterate through the think-confirm stages to ensure the verified rationale is consistent with the answer. We conduct experiments on a range of knowledge-based visual reasoning datasets. We found our VCTP enjoys several benefits, 1). it achieves better performance than the previous few-shot learning baselines; 2). it enjoys the total transparency and trustworthiness of the whole reasoning process by providing rationales for each reasoning step; 3). it is computation-efficient compared with other fine-tuning baselines. Our code is available at https://github.com/UMass-Foundation-Model/VisualCoT.git

Leveraging Text-to-Text Pretrained Language Models for Question Answering in Chemistry.

In this study, we present a question answering (QA) system for chemistry, named Marie, with the use of a text-to-text pretrained language model to attain accurate data retrieval. The underlying data store is "The World Avatar" (TWA), a general world model consisting of a knowledge graph that evolves over time. TWA includes information about chemical species such as their chemical and physical properties, applications, and chemical classifications. Building upon our previous work on KGQA for chemistry, this advanced version of Marie leverages a fine-tuned Flan-T5 model to seamlessly translate natural language questions into SPARQL queries with no separate components for entity and relation linking. The developed QA system demonstrates competence in providing accurate results for complex queries that involve many relation hops as well as showcasing the ability to balance correctness and speed for real-world usage. This new approach offers significant advantages over the prior implementation that relied on knowledge graph embedding. Specifically, the updated system boasts high accuracy and great flexibility in accommodating changes and evolution of the data stored in the knowledge graph without necessitating retraining. Our evaluation results underscore the efficacy of the improved system, highlighting its superior accuracy and the ability in answering complex questions compared to its predecessor.

"Your Turn!": Using Finger Pointing and PALM-UP Actions to Ask Questions in Norwegian Sign Language

Abstract: Researchers examining the structure of questions in signed languages, often using elicited data from informants, have proposed that there are specific manual and nonmanual actions produced by signers to indicate different question types (e.g., Zeshan 2004), for example, raised eyebrows for polar questions. In the current study, we add to this body of work with an investigation into the use of two common interactional behaviors—finger pointing and palm-up actions—as they occur in questions asked during naturalistic conversations in Norwegian Sign Language. The interactional work finger pointing and palm-up actions can achieve in various settings is well-documented (e.g., Bavelas 1994), and their presence in question contexts has been observed in Norwegian Sign Language (Ferrara 2020). In total, 455 questions produced by thirty-four deaf signers were examined as they appeared across dyadic, triadic, and multiparty conversations recorded in the Norwegian Sign Language Corpus (Ferrara and Ringsø 2021; Ferrara and Bø 2022; Ferrara in prep). These questions were then examined for the presence of interactional finger pointing and palm-up actions. Analyses firstly revealed that interactional finger pointing and palm-up actions appeared in many of the questions examined. Within this subset of the data, further analysis showed that these forms may be the only manual action that signals an utterance as a question, or they may occur in combination with other manual (e.g., wh-question forms: how, where) and nonmanual actions (e.g., furrowed or raised brows). In many of these questions, the two forms appear in utterance-final position, thereby indexing an interlocutor in order to solicit a response. Such initial observations suggest that these two manual actions must be considered as part of the constellation of behaviors that characterize the highly indexical and interactional nature of question-asking in Norwegian Sign Language.

Speech-to-SQL: toward speech-driven SQL query generation from natural language question

Speech-to-SQL: toward speech-driven SQL query generation from natural language question

A Relation Embedding Assistance Networks for Multi-hop Question Answering

Multi-hop Knowledge Graph Question Answering aims at finding an entity to answer natural language questions from knowledge graphs. When humans perform multi-hop reasoning, people tend to focus on specific relations across different hops and confirm the next entity. Therefore, most algorithms choose the wrong specific relation, which makes the system deviate from the correct reasoning path. The specific relation at each hop plays an important role in multi-hop question answering. Existing work mainly relies on the question representation as relation information, which cannot accurately calculate the specific relation distribution. In this article, we propose an interpretable assistance framework that fully utilizes the relation embeddings to assist in calculating relation distributions at each hop. Moreover, we employ the fusion attention mechanism to ensure the integrity of relation information and hence to enrich the relation embeddings. The experimental results on three English datasets and one Chinese dataset demonstrate that our method significantly outperforms all baselines. The source code of REAN will be available at https://github.com/2399240664/REAN

Leveraging large language models for predictive chemistry

Machine learning has transformed many fields and has recently found applications in chemistry and materials science. The small datasets commonly found in chemistry sparked the development of sophisticated machine learning approaches that incorporate chemical knowledge for each application and, therefore, require specialized expertise to develop. Here we show that GPT-3, a large language model trained on vast amounts of text extracted from the Internet, can easily be adapted to solve various tasks in chemistry and materials science by fine-tuning it to answer chemical questions in natural language with the correct answer. We compared this approach with dedicated machine learning models for many applications spanning the properties of molecules and materials to the yield of chemical reactions. Surprisingly, our fine-tuned version of GPT-3 can perform comparably to or even outperform conventional machine learning techniques, in particular in the low-data limit. In addition, we can perform inverse design by simply inverting the questions. The ease of use and high performance, especially for small datasets, can impact the fundamental approach to using machine learning in the chemical and material sciences. In addition to a literature search, querying a pre-trained large language model might become a routine way to bootstrap a project by leveraging the collective knowledge encoded in these foundation models, or to provide a baseline for predictive tasks.

DocFlow: A Visual Analytics System for Question-Based Document Retrieval and Categorization.

A systematic review (SR) is essential with up-to-date research evidence to support clinical decisions and practices. However, the growing literature volume makes it challenging for SR reviewers and clinicians to discover useful information efficiently. Many human-in-the-loop information retrieval approaches (HIR) have been proposed to rank documents semantically similar to users' queries and provide interactive visualizations to facilitate document retrieval. Given that the queries are mainly composed of keywords and keyphrases retrieving documents that are semantically similar to a query does not necessarily respond to the clinician's need. Clinicians still have to review many documents to find the solution. The problem motivates us to develop a visual analytics system, DocFlow, to facilitate information-seeking. One of the features of our DocFlow is accepting natural language questions. The detailed description enables retrieving documents that can answer users' questions. Additionally, clinicians often categorize documents based on their backgrounds and with different purposes (e.g., populations, treatments). Since the criteria are unknown and cannot be pre-defined in advance, existing methods can only achieve categorization by considering the entire information in documents. In contrast, by locating answers in each document, our DocFlow can intelligently categorize documents based on users' questions. The second feature of our DocFlow is a flexible interface where users can arrange a sequence of questions to customize their rules for document retrieval and categorization. The two features of this visual analytics system support a flexible information-seeking process. The case studies and the feedback from domain experts demonstrate the usefulness and effectiveness of our DocFlow.

Analysis of Arabic Language Questions Based on Higher Order Thinking Skills (HOTS) in the Arabic Language Book for Class XI Ministry of Religion

Analysis of Arabic Language Questions Based on Higher Order Thinking Skills (HOTS) in the Arabic Language Book for Class XI Ministry of Religion

A Dual-Attention Learning Network With Word and Sentence Embedding for Medical Visual Question Answering.

Research in medical visual question answering (MVQA) can contribute to the development of computer-aided diagnosis. MVQA is a task that aims to predict accurate and convincing answers based on given medical images and associated natural language questions. This task requires extracting medical knowledge-rich feature content and making fine-grained understandings of them. Therefore, constructing an effective feature extraction and understanding scheme are keys to modeling. Existing MVQA question extraction schemes mainly focus on word information, ignoring medical information in the text, such as medical concepts and domain-specific terms. Meanwhile, some visual and textual feature understanding schemes cannot effectively capture the correlation between regions and keywords for reasonable visual reasoning. In this study, a dual-attention learning network with word and sentence embedding (DALNet-WSE) is proposed. We design a module, transformer with sentence embedding (TSE), to extract a double embedding representation of questions containing keywords and medical information. A dual-attention learning (DAL) module consisting of self-attention and guided attention is proposed to model intensive intramodal and intermodal interactions. With multiple DAL modules (DALs), learning visual and textual co-attention can increase the granularity of understanding and improve visual reasoning. Experimental results on the ImageCLEF 2019 VQA-MED (VQA-MED 2019) and VQA-RAD datasets demonstrate that our proposed method outperforms previous state-of-the-art methods. According to the ablation studies and Grad-CAM maps, DALNet-WSE can extract rich textual information and has strong visual reasoning ability.

HFD: Hierarchical feature decoupling for SQL generation from text

Text-to-SQL, a computational linguistics task, seeks to facilitate the conversion of natural language queries into SQL queries. Recent methodologies have leveraged the concept of slot-filling in conjunction with predetermined SQL templates to effectively bridge the semantic gap between natural language questions and structured database queries, achieving commendable performance by harnessing the power of multi-task learning. However, employing identical features across diverse tasks is an ill-suited practice, fraught with inherent drawbacks. Firstly, based on our observation, there are clear boundaries in the natural language corresponding to SELECT and WHERE clauses. Secondly, the exclusive features integral to each subtask are inadequately emphasized and underutilized, thereby hampering the acquisition of discriminative features for each specific subtask. In an endeavor to rectify these issues, the present work introduces an innovative approach: the hierarchical feature decoupling model for SQL query generation from natural language. This novel approach involves the deliberate separation of features pertaining to subtasks within both SELECT and WHERE clauses, further dissociating these features at the subtask level to foster better model performance. Empirical results derived from experiments conducted on the WikiSQL benchmark dataset reveal the superiority of the proposed approach over several state-of-the-art baseline methods in the context of text-to-SQL query generation.

Grand entretien avec François Laplantine

ObjectiveBased on an interview with François Laplantine, anthropologist and professor emeritus at the University Lumière Lyon 2, the aim is to explore the main milestones in his career, from the philosophy of evil to the anthropology of illness and ethnopsychiatry, in particular psychoanalytic ethnopsychiatry, then from work on Brazil and Japan to a focus on the theme of artistic creation. MethodThe method consists of an interview with François Laplantine. The questions deal first with his personal background, then with the reasons for the evolution of the themes on which his works are based, and finally with his exploration of various fields of creation: literature, cinema, theatre and contemporary dance. ResultsFrançois Laplantine describes his transition from the philosophy of evil, with his first thesis in philosophy under the supervision of Paul Ricœur, to the anthropology of illness, with his encounter with Devereux. He describes his work on the question of métissages, then his transition from psychoanalysis to an initiation into Candomblé in Brazil, a practice of the multiple that he likens to different forms of theatre in France. He then traces the evolution of his writings from his explorations in Brazil and Japan to artistic creation, particularly literature, music, dance, and film. He discusses the work of many artists, as well as different types of staging of disasters, particularly minimalist stagings in Japanese theatre, with nō and butô. The question of language, of the unspeakable and the invisible, appears central to his work in relation to creation. He also discusses his work on the imaginary and the off-screen, as well as on time and space, in relation to the issue of creation. DiscussionFrançois Laplantine defends the idea that one of the missions of contemporary arts is to attempt to express catastrophe; he explores the limits of language and argues that one of the functions of theatre, and even more so of contemporary dance, is to stage the drama of the inadequacy of language. Recourse to artistic creation is a necessity, in the tension between telling and showing. ConclusionExploring artistic creation appears to be a royal road for an anthropologist like François Laplantine.

Review of the Cinematic Elements in “Two Men and A Wardrobe”

The present research focuses on Roman Polanski’s short film “Two Men and a Wardrobe,” unraveling the sophisticated cinematic language, symbolic motifs, and societal commentary throughout the plot. The short film is one of the director’s earliest works and yet one of the most evocative of the society that he lived in. The study analyzes Polanski’s directorial approach, distinguished by his careful use of simple yet conspicuous settings and recurring motifs, to portray themes of loneliness, surrealism, and resilience within the framework of the cityscape and how townspeople treat strangers. Several stylistic and language questions that Polansky uses are explored. First, the research reveals the director’s expert use of lighting to elicit resonance, particularly in emphasizing the characters’ experiences and the film’s overall mood. It also investigates the film’s relationship stories of Holocaust suffering and its portrayal of societal difficulties, in line with the cinematic discourse around films such as “La Vita è Bella” and “Schindler’s List.” The study provides vital insights into the tremendous effect of Polanski’s work by exploring the theme exploration and artistic choices in “Two Men and a Wardrobe,” clarifying its significance within the fields of film and societal reflection.

Creation and importance of language corps in Uzbekistan

The article discusses the transformation of language into the language of the Internet, computer technology, mathematical linguistics, its continuation and the formation and development of computer linguistics, in particular the question of modeling natural languages for artificial intelligence. The Uzbek National Corps plays an important role in enhancing the international status of the Uzbek language. The work carried out in the field of computer linguistics plays an important role in resolving existing problems in the Uzbek language. The question of the linguistic and extralinguistic separation of special tags for marking texts and their components is studied in particular.The coding requirements for important text information are defined. The state analyzes the linguistic module and the algorithm and its types from independent components of the linguistic program code. The need for algorithms for phonological, morphological and spelling rules for the formation of the lexical and grammatical code is scientifically substantiated. The importance of such linguistic modules as phonology, morphology and spelling in the formation of the linguistic base of the national corpus of the Uzbek language is emphasized. The article examines the corpus’s primary purpose as a complex linguistic source, as well as the fact that it primarily contains two sorts of information and its types. The key effective capabilities of the corpus, according to the paper, are reducing time spent on the text analysis process and being able to explain the properties of language units in speech with thousands of instances. The national corpus, the educational corpus, and the parallel corpus are all discussed in the subject of computer linguistics. It was stressed that linguistic and extralinguistic tagging of them, the development of corpus formation algorithms, and the establishment of corpus linguistic support are all societal need. It recognizes the urgency of developing the basis for the creation of the Uzbek language corpus, conducting research in the field of computer linguistics as a scientific and theoretical source.

Intelligent Problem Solver in Database Systems based on Ontology Integration through Text-to-SQL

The knowledge of courses can be represented by using ontology to create intelligent educational systems. This study proposes the Onto-Linking model as a knowledge framework that expresses the knowledge of the inputted schema to investigate the schema linking problem of the Text-to-SQL model. It combines the ontology with the structure of the schema. The proposed ontology is utilized to encapsulate the semantics of the intellectual elements of the schema, such as the table names, column names, foreignprimary key restrictions, and information about the probing schema connection. Therefore, the model makes it easier to accurately translate natural language questions into SQL queries. It improves query creation, helps with error handling, and supports query validation by helping the model better grasp the query's intent. The outcomes of the pedagogically oriented model aimed at guiding learners to comprehend the process of reasoning to attain the respective solution.

A Comprehensive Evaluation of Neural SPARQL Query Generation From Natural Language Questions

A Comprehensive Evaluation of Neural SPARQL Query Generation From Natural Language Questions

Student Teachers’ Practices: A Case of English Language Teaching Classrooms

The primary purpose of carrying out the study was to explore English language teaching (ELT) practices adopted by the student teachers (STs) in their classrooms and to examine the effectiveness of teaching practice (TP) of Janapriya Multiple Campus (JMC), Pokhara Nepal. The information was collected through unstructured tools such as observation, questionnaire, and focused group discussion (FGD). The collected information was analyzed through qualitative methods. The STs were found punctual, dedicated, and responsible in their TP. They were found practicing various classroom activities such as language games, role plays, discussions, group works, pair works, translations, question and answer, and drillings during their TP. In addition, they also were found conducting various activities in the classrooms such as debates, storytelling, essay writing, quiz contests, and eloquence. Moreover, they also carried out a case study and wrote a report, prepared a Secondary Education Examination (SEE) model question, constructed various types of questions, observed three peers’ classes, and kept a record of their classroom activities and teaching materials. Completing all the required activities, the STs performed effectively during their TP. The findings of the study can be useful to make the TP more effective.

Jhumpa Lahiri and the Translation Memoir: To Write and Exist Beyond the Mother Tongue

ABSTRACT In 2016, following the publication of her memoir, In Other Words, Pulitzer Prize author Jhumpa Lahiri announced her intention to abandon English to solely write in Italian. Six years later, having become an acclaimed Italian writer and translator, Lahiri returns to English to consider her role as a ‘postmonolingual’ subject (Yildiz) in an era still dominated by a paradigmatic monolingualism. Released in 2022, Translating Myself and Others can be considered to all effects Lahiri’s translation memoir as it extends to the field of translation, the reflections initiated by In Other Words on the questions of language, identity, and cultural assignation. Through the analysis of both texts, this article discusses how the notion of ‘mother tongue’ is progressively deconstructed in Lahiri’s memoirs and how translation and self-translation become the vehicle through which overcoming what Yasemin Yildiz defines as the ‘monolingual paradigm’.

Implementation of Recurrent Neural Network (RNN) for Question Similarity Identification in Indonesian Language

In a question-and-answer forum, the identification of question similarity is used to determine how similar two questions are. This procedure makes sure that user-submitted questions are compared to the questions in a database for matches to improve system performance on the online Q&A platform. Currently, question similarity is mostly done in foreign languages. The purpose of this research is to identify question similarities and evaluate the effectiveness of the methods used in Indonesian language questions. The data used is a public dataset with labeled pairs of questions as 0 and 1 where label 0 for different pairs of questions and label 1 for the same pairs of questions. The method used is a Recurrent Neural Network (RNN) with the Manhattan Distance approach to calculate the similarity distance between two questions. The question pairs are taken as two inputs with a reference label to identify the similarity distance between the two question inputs. We evaluated the model using three different optimizers namely RMSprop, Adam, and Adagrad. The best results were obtained using the Adam optimizer with 80:20 ratio split-data and overall accuracy is 76%, precision is 74%, recall is 98.8%, and F1-score is 85.1%.

Geographic Knowledge Base Question Answering over OpenStreetMap

In recent years, question answering on knowledge bases (KBQA) has emerged as a promising approach for providing unified, user-friendly access to knowledge bases. Nevertheless, existing KBQA systems struggle to answer spatial-related questions, prompting the introduction of geographic knowledge ba se question answering (GeoKBQA) to address such challenges. Current GeoKBQA systems face three primary issues: (1) the limited scale of questions, restricting the effective application of neural networks; (2) reliance on rule-based approaches dependent on predefined templates, resulting in coverage and scalability challenges; and (3) the assumption of the availability of a golden entity, limiting the practicality of GeoKBQA systems. In this work, we aim to address these three critical issues to develop a practical GeoKBQA system. We construct a large-scale, high-quality GeoKBQA dataset and link mentions in the questions to entities in OpenStreetMap using an end-to-end entity-linking method. Additionally, we develop a query generator that translates natural language questions, along with the entities predicted by entity linking into corresponding GeoSPARQL queries. To the best of our knowledge, this work presents the first purely neural-based GeoKBQA system with potential for real-world application.

Feature-driven new sound category formation: computational implementation with the L2LP model and beyond

One of the primary questions of second language (L2) acquisition research is how a new sound category is formed to allow for an L2 contrast that does not exist in the learner's first language (L1). Most models rely crucially on perceived (dis)similarities between L1 and L2 sounds, but a precise definition of what constitutes “similarity” has long proven elusive. The current study proposes that perceived cross-linguistic similarities are based on feature-level representations, not segmental categories. We investigate how L1 Japanese listeners learn to establish a new category for L2 American English /æ/ through a perception experiment and computational, phonological modeling. Our experimental results reveal that intermediate-level Japanese learners of English perceive /æ/ as an unusually fronted deviant of Japanese /a/. We implemented two versions of the Second Language Linguistic Perception (L2LP) model with Stochastic Optimality Theory—one mapping acoustic cues to segmental categories and another to features—and compared their simulated learning results to the experimental results. The segmental model was theoretically inadequate as it was unable explain how L1 Japanese listeners notice the deviance of /æ/ from /a/ in the first place, and was also practically implausible because the predicted overall perception patterns were too native English-like compared to real learners' perception. The featural model, however, showed that the deviance of /æ/ could be perceived due to an ill-formed combination of height and backness features, namely */low, front/. The featural model, therefore, reflected the experimental results more closely, where a new category was formed for /æ/ but not for other L2 vowels /ɛ/, /ʌ/, and /ɑ/, which although acoustically deviate from L1 /e/, /a/, and /o/, are nonetheless featurally well-formed in L1 Japanese, namely /mid, front/, /low, central/, and /mid, back/. The benefits of a feature-based approach for L2LP and other L2 models, as well as future directions for extending the approach, are discussed.