Search

BACKGROUND: Although uncertainties exist regarding implementation, artificial intelligence-driven generative language models (GLMs) have enormous potential in medicine. Deployment of GLMs could improve patient comprehension of clinical texts and improve low health literacy. OBJECTIVE: The goal of this study is to evaluate the potential of ChatGPT-3.5 and GPT-4 to tailor the complexity of medical information to patient-specific input education level, which is crucial if it is to serve as a tool in addressing low health literacy. METHODS: Input templates related to 2 prevalent chronic diseases-type II diabetes and hypertension-were designed. Each clinical vignette was adjusted for hypothetical patient education levels to evaluate output personalization. To assess the success of a GLM (GPT-3.5 and GPT-4) in tailoring output writing, the readability of pre- and posttransformation outputs were quantified using the Flesch reading ease score (FKRE) and the Flesch-Kincaid grade level (FKGL). RESULTS: Responses (n=80) were generated using GPT-3.5 and GPT-4 across 2 clinical vignettes. For GPT-3.5, FKRE means were 57.75 (SD 4.75), 51.28 (SD 5.14), 32.28 (SD 4.52), and 28.31 (SD 5.22) for 6th grade, 8th grade, high school, and bachelors, respectively; FKGL mean scores were 9.08 (SD 0.90), 10.27 (SD 1.06), 13.4 (SD 0.80), and 13.74 (SD 1.18). GPT-3.5 only aligned with the prespecified education levels at the bachelors degree. Conversely, GPT-4s FKRE mean scores were 74.54 (SD 2.6), 71.25 (SD 4.96), 47.61 (SD 6.13), and 13.71 (SD 5.77), with FKGL mean scores of 6.3 (SD 0.73), 6.7 (SD 1.11), 11.09 (SD 1.26), and 17.03 (SD 1.11) for the same respective education levels. GPT-4 met the target readability for all groups except the 6th-grade FKRE average. Both GLMs produced outputs with statistically significant differences (P

BACKGROUND: Pathology reports contain key information about the patients diagnosis as well as important gross and microscopic findings. These information-rich clinical reports offer an invaluable resource for clinical studies, but data extraction and analysis from such unstructured texts is often manual and tedious. While neural information retrieval systems (typically implemented as deep learning methods for natural language processing) are automatic and flexible, they typically require a large domain-specific text corpus for training, making them infeasible for many medical subdomains. Thus, an automated data extraction method for pathology reports that does not require a large training corpus would be of significant value and utility. OBJECTIVE: To develop a language model-based neural information retrieval system that can be trained on small datasets and validate it by training it on renal transplant-pathology reports to extract relevant information for two predefined questions: (1) What kind of rejection does the patient show?; (2) What is the grade of interstitial fibrosis and tubular atrophy (IFTA)? METHODS: Kidney BERT was developed by pre-training Clinical BERT on 3.4K renal transplant pathology reports and 1.5M words. Then, exKidneyBERT was developed by extending Clinical BERTs tokenizer with six technical keywords and repeating the pre-training procedure. This extended the models vocabulary. All three models were fine-tuned with information retrieval heads. RESULTS: The model with extended vocabulary, exKidneyBERT, outperformed Clinical BERT and Kidney BERT in both questions. For rejection, exKidneyBERT achieved an 83.3% overlap ratio for antibody-mediated rejection (ABMR) and 79.2% for T-cell mediated rejection (TCMR). For IFTA, exKidneyBERT had a 95.8% exact match rate. CONCLUSION: ExKidneyBERT is a high-performing model for extracting information from renal pathology reports. Additional pre-training of BERT language models on specialized small domains does not necessarily improve performance. Extending the BERT tokenizers vocabulary library is essential for specialized domains to improve performance, especially when pre-training on small corpora.

The digitization of operation and maintenance in the intelligent power grid equipment relies on a diverse array of information for smart decision-making. In the domain of intelligent decision generation, proficiency is contingent upon extensive learning from copious amounts of text. This necessitates not only robust processing capabilities but also a high level of specialization. In addressing situations where authorization is lacking, pre-trained language models (PLMs) have already provided ideas when confronted with specialized domains or tasks. In consideration of the complexity of textual content in the field of the power grid, which encompasses a multitude of specialized knowledge and involves an abundance of proprietary terminology, we have undertaken an exploration of pre-trained model specialization using the power grid domain as an example, specifically for the task of generating maintenance strategies. A two-stage fine-tuning approach (P2FT) is employed, utilizing a large-scale pre-training model specifically designed for natural language processing. The efficacy and practical value of this method were evaluated through multiple metrics, juxtaposed with other advanced approaches involving low-parameter or parameter-free fine-tuning methods. Through a meticulous analysis and validation of experimental outcomes, we have corroborated the feasibility and practical application value of employing this approach for pre-trained model specialization. Additionally, it has furnished valuable guidance for text generation within both the Chinese language domain and the power grid domain. [ABSTRACT FROM AUTHOR]

This paper proposes an approach to sentiment analysis of student evaluation feedback using transformer-based language models. The primary objective of this study is to conduct an in-depth analysis of sentiment expressed in student evaluation feedback, with a focus on introducing contextual understanding into the sentiment classification process. In this research, four different variants of transformer language models were assessed, namely multilingual bidirectional encoder representations from transformers (MBERT), IndoBERT, RoBERTa Indonesia, and generative pre-trained transformer (GPT-2 Indonesia). Additionally, we also compared the performance of transformer models with two traditional models, namely support vector machine (SVM) and Naive Bayes (NB). The evaluation was conducted using feedback data collected from the Evaluasi Dosen oleh Mahasiswa (EDOM) system at Riau University, which had been categorized as either positive or negative. The outcomes indicate that IndoBERT base uncased exhibits the highest performance, with precision, accuracy, and recall values of 0.858, 0.929, and 0.911, respectively. This observation highlights the effectiveness of transformer-based language models in sentiment analysis of student evaluation feedback and provides insights for improving educational assessment practices. [ABSTRACT FROM AUTHOR]

Drug resistance poses a crucial challenge in healthcare, with response rates to chemotherapy and targeted therapy remaining low. Individual patient's resistance is exacerbated by the intricate heterogeneity of tumor cells, presenting significant obstacles to effective treatment. To address this challenge, DrugFormer, a novel graph‐augmented large language model designed to predict drug resistance at single‐cell level is proposed. DrugFormer integrates both serialized gene tokens and gene‐based knowledge graphs for the accurate predictions of drug response. After training on comprehensive single‐cell data with drug response information, DrugFormer model presents outperformance, with higher F1, precision, and recall in predicting drug response. Based on the scRNA‐seq data from refractory multiple myeloma (MM) and acute myeloid leukemia (AML) patients, DrugFormer demonstrates high efficacy in identifying resistant cells and uncovering underlying molecular mechanisms. Through pseudotime trajectory analysisunique drug‐resistant cellular states associated with poor patient outcomes are revealed. Furthermore, DrugFormer identifies potential therapeutic targets, such as COX8A, for overcoming drug resistance across different cancer types. In conclusion, DrugFormer represents a significant advancement in the field of drug resistance prediction, offering a powerful tool for unraveling the heterogeneity of cellular response to drugs and guiding personalized treatment strategies. [ABSTRACT FROM AUTHOR]

User review texts provide valuable information for recommender systems, as they express various dimensions and perspectives regarding the experience of a user with a specific item. Consequently, many studies have proposed recommender systems based on review texts. However, because review texts typically contain a high proportion of noise that is not related to user preferences or item characteristics, existing studies that input the entire review text into the model are vulnerable to noise issues. Therefore, this study proposes a methodology for extracting key attributes based on in-context learning(ICL) to fundamentally address the noise problem in review texts. We used zero-shot, one-shot, and few-shot large language model (LLM) ICL to generate key attributes that define user preferences and item characteristics from integrated review texts, and we trained a recommender system to predict user ratings on items using the generated key attributes as new input. Our proposed research is the first to create and utilize new user and item characteristics through LLM ICL for a recommender system. Experiments demonstrate that our methodology effectively generates key attributes related to user preferences and item characteristics from review texts and achieves superior predictive performance compared to existing review-based recommender systems. [ABSTRACT FROM AUTHOR]

Combinatorial nature of polymer matrix composites design requires a robust predictive model to accurately predict the mechanical properties of polymer composites, thereby reducing the need for extensive and costly trial-and-error approaches in their manufacturing. However, traditional prediction models have been either lacking in accuracy or too resource-intensive for practical use. This study proposes an advanced Transformer-based predictive model simultaneously considering various variables that can influence mechanical properties, while utilizing only a minimal amount of training data. In developing this model, we utilize an extensive dataset across 294 types of polymer composites, using a diverse range of polymers and reinforcements, providing a comprehensive basis for the model's predictions. The model employs a Transformer-based transfer learning technique, known for its efficiency with small datasets, to predict essential mechanical properties such as tensile strength, tensile modulus, flexural strength, flexural modulus and density. It shows high predictive accuracy (R2 = 92%) and makes reliable predictions for combinations of polymer composites that have not been trained on (R2 = 82%). Additionally, the model's effectiveness and learning process are validated through Explainable Artificial Intelligence analysis and latent space visualization. [ABSTRACT FROM AUTHOR]

Despite decades of antibody research, it remains challenging to predict the specificity of an antibody solely based on its sequence. Two major obstacles are the lack of appropriate models and the inaccessibility of datasets for model training. In this study, we curated >5,000 influenza hemagglutinin (HA) antibodies by mining research publications and patents, which revealed many distinct sequence features between antibodies to HA head and stem domains. We then leveraged this dataset to develop a lightweight memory B cell language model (mBLM) for sequence-based antibody specificity prediction. Model explainability analysis showed that mBLM could identify key sequence features of HA stem antibodies. Additionally, by applying mBLM to HA antibodies with unknown epitopes, we discovered and experimentally validated many HA stem antibodies. Overall, this study not only advances our molecular understanding of the antibody response to the influenza virus but also provides a valuable resource for applying deep learning to antibody research. • Assembled a dataset of 5,561 published antibodies to influenza HA from 132 donors • Antibodies to HA head and stem domains have distinct convergent sequence features • Developed a lightweight language model (mBLM) for antibody specificity prediction • Discovered HA stem antibodies and key somatic hypermutations using mBLM Predicting antibody specificity based solely on sequence has remained an obstacle in humoral research. Leveraging a curated dataset of >5,000 influenza hemagglutinin (HA) antibodies, Wang et al. develop a lightweight memory B cell language model for antibody specificity prediction. This model identifies unique HA stem antibodies and key antibody sequence features. [ABSTRACT FROM AUTHOR]

Reports of Large Language Models (LLMs) passing board examinations have spurred medical enthusiasm for their clinical integration. Through a narrative review, we reflect upon the skill shifts necessary for clinicians to succeed in an LLM-enabled world, achieving benefits while minimizing risks. We suggest how medical education must evolve to prepare clinicians capable of navigating human-AI systems. [ABSTRACT FROM AUTHOR]

Background: As large language models continue to expand in size and diversity, their substantial potential and the relevance of their applications are increasingly being acknowledged. The rapid advancement of these models also holds profound implications for the long-term design of stimulus-responsive materials used in drug delivery. Methods: The large model used Hugging Face's Transformers package with BigBird, Gemma, and GPT NeoX architectures. Pre-training used the PubChem dataset, and fine-tuning used QM7b. Chemist instruction training was based on Direct Preference Optimization. Drug Likeness, Synthetic Accessibility, and PageRank Scores were used to filter molecules. All computational chemistry simulations were performed using ORCA and Time-Dependent Density-Functional Theory. Results: To optimize large models for extensive dataset processing and comprehensive learning akin to a chemist's intuition, the integration of deeper chemical insights is imperative. Our study initially compared the performance of BigBird, Gemma, GPT NeoX, and others, specifically focusing on the design of photoresponsive drug delivery molecules. We gathered excitation energy data through computational chemistry tools and further investigated light-driven isomerization reactions as a critical mechanism in drug delivery. Additionally, we explored the effectiveness of incorporating human feedback into reinforcement learning to imbue large models with chemical intuition, enhancing their understanding of relationships involving -N=N- groups in the photoisomerization transitions of photoresponsive molecules. Conclusions: We implemented an efficient design process based on structural knowledge and data, driven by large language model technology, to obtain a candidate dataset of specific photoswitchable molecules. However, the lack of specialized domain datasets remains a challenge for maximizing model performance. [ABSTRACT FROM AUTHOR]

Copyright of Przegląd Elektrotechniczny is the property of Przeglad Elektrotechniczny and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Objectives To develop and validate a novel measure, action entropy, for assessing the cognitive effort associated with electronic health record (EHR)-based work activities. Materials and Methods EHR-based audit logs of attending physicians and advanced practice providers (APPs) from four surgical intensive care units in 2019 were included. Neural language models (LMs) were trained and validated separately for attendings' and APPs' action sequences. Action entropy was calculated as the cross-entropy associated with the predicted probability of the next action, based on prior actions. To validate the measure, a matched pairs study was conducted to assess the difference in action entropy during known high cognitive effort scenarios, namely, attention switching between patients and to or from the EHR inbox. Results Sixty-five clinicians performing 5 904 429 EHR-based audit log actions on 8956 unique patients were included. All attention switching scenarios were associated with a higher action entropy compared to non-switching scenarios (P  < .001), except for the from-inbox switching scenario among APPs. The highest difference among attendings was for the from-inbox attention switching: Action entropy was 1.288 (95% CI, 1.256-1.320) standard deviations (SDs) higher for switching compared to non-switching scenarios. For APPs, the highest difference was for the to-inbox switching, where action entropy was 2.354 (95% CI, 2.311-2.397) SDs higher for switching compared to non-switching scenarios. Discussion We developed a LM-based metric, action entropy, for assessing cognitive burden associated with EHR-based actions. The metric showed discriminant validity and statistical significance when evaluated against known situations of high cognitive effort (ie, attention switching). With additional validation, this metric can potentially be used as a screening tool for assessing behavioral action phenotypes that are associated with higher cognitive burden. Conclusion An LM-based action entropy metric—relying on sequences of EHR actions—offers opportunities for assessing cognitive effort in EHR-based workflows. [ABSTRACT FROM AUTHOR]

A crucial step in understanding natural language is detecting mentions that refer to real-world entities in a text and correctly identifying their boundaries. Mention detection is commonly considered a preprocessing step in coreference resolution which is shown to be helpful in several language processing applications such as machine translation and text summarization. Despite recent efforts on Turkish coreference resolution, no standalone neural solution to mention detection has been proposed yet. In this article, we present two models designed for detecting Turkish mentions by using feed-forward neural networks. Both models extract all spans up to a fixed length from input text as candidates and classify them as mentions or not mentions. The models differ in terms of how candidate text spans are represented. The first model represents a span by focusing on its first and last words, whereas the representation also covers the preceding and proceeding words of a span in the second model. Mention span representations are formed by using contextual embeddings, part-of-speech embeddings, and named-entity embeddings of words in interest where contextual embeddings are obtained from pretrained Turkish language models. In our evaluation studies, we not only assess the impact of mention representation strategies on system performance but also demonstrate the usability of different pretrained language models in resolution task. We argue that our work provides useful insights to the existing literature and the first step in understanding the effectiveness of neural architectures in Turkish mention detection. [ABSTRACT FROM AUTHOR]

ChatGPT is a language model that was trained on a large dataset including medical literature. Several studies have described the performance of ChatGPT on medical exams. In this study, we examine its performance in answering factual knowledge questions regarding clinical pharmacy. Questions were obtained from a Dutch application that features multiple‐choice questions to maintain a basic knowledge level for clinical pharmacists. In total, 264 clinical pharmacy‐related questions were presented to ChatGPT and responses were evaluated for accuracy, concordance, quality of the substantiation, and reproducibility. Accuracy was defined as the correctness of the answer, and results were compared to the overall score by pharmacists over 2022. Responses were marked concordant if no contradictions were present. The quality of the substantiation was graded by two independent pharmacists using a 4‐point scale. Reproducibility was established by presenting questions multiple times and on various days. ChatGPT yielded accurate responses for 79% of the questions, surpassing pharmacists' accuracy of 66%. Concordance was 95%, and the quality of the substantiation was deemed good or excellent for 73% of the questions. Reproducibility was consistently high, both within day and between days (>92%), as well as across different users. ChatGPT demonstrated a higher accuracy and reproducibility to factual knowledge questions related to clinical pharmacy practice than pharmacists. Consequently, we posit that ChatGPT could serve as a valuable resource to pharmacists. We hope the technology will further improve, which may lead to enhanced future performance. [ABSTRACT FROM AUTHOR]

The study discusses modern approaches to generating poetic text. It carries out a comparative analysis of various language models and proposes and implements the architecture of natural language processing—a pipeline for the automatic generation of meaningful poetic text on the basis of a language model, genetic algorithm, and classifier of semantic text coherence. [ABSTRACT FROM AUTHOR]

Copyright of Procesamiento del Lenguaje Natural is the property of Sociedad Espanola para el Procesamiento del Lenguaje Natural and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Traditionally, either applying the hard prompt for sentences by handcrafting the prompt templates or directly optimizing the soft or continuous prompt may not sufficiently generalize for unseen domain data. This paper presents a parameter efficient learning for domain-agnostic soft prompt which is developed for few-shot unsupervised domain adaptation. A pre-trained language model (PLM) is frozen and utilized to extract knowledge for unseen domains in various language understanding tasks. The meta learning and optimization over a set of trainable soft tokens is performed by minimizing the cross-entropy loss for masked language model from support and query data in source and target domains, respectively, where the masked tokens for text category and random masking are predicted. The meta soft prompt is learned through a doublylooped optimization for individual learners and a meta learner when implementing the unsupervised domain adaptation. The PLM is then closely adapted to compensate the domain shift in a target domain. The domain adaptation loss and the prompt-based classification loss are jointly minimized through meta learning. The experiments on multi-domain natural language understanding illustrate the merit of the proposed meta soft prompt in pre-trained language modeling under few-shot setting. [ABSTRACT FROM AUTHOR]

Integrating deep learning techniques, particularly language models (LMs), with knowledge representation techniques like ontologies has raised widespread attention, urging the need of a platform that supports both paradigms. Although packages such as OWL API and Jena offer robust support for basic ontology processing features, they lack the capability to transform various types of information within ontologies into formats suitable for downstream deep learning-based applications. Moreover, widely-used ontology APIs are primarily Java-based while deep learning frameworks like PyTorch and Tensorflow are mainly for Python programming. To address the needs, we present DeepOnto , a Python package designed for ontology engineering with deep learning. The package encompasses a core ontology processing module founded on the widely-recognised and reliable OWL API, encapsulating its fundamental features in a more "Pythonic" manner and extending its capabilities to incorporate other essential components including reasoning, verbalisation, normalisation, taxonomy, projection, and more. Building on this module, DeepOnto offers a suite of tools, resources, and algorithms that support various ontology engineering tasks, such as ontology alignment and completion, by harnessing deep learning methods, primarily pre-trained LMs. In this paper, we also demonstrate the practical utility of DeepOnto through two use-cases: the Digital Health Coaching in Samsung Research UK and the Bio-ML track of the Ontology Alignment Evaluation Initiative (OAEI). [ABSTRACT FROM AUTHOR]

Multi-target multi-camera tracking is crucial to intelligent transportation systems. Numerous recent studies have been undertaken to address this issue. Nevertheless, using the approaches in real-world situations is challenging due to the scarcity of publicly available data and the laborious process of manually annotating the new dataset and creating a tailored rule-based matching system for each camera scenario. To address this issue, we present a novel solution termed LaMMOn, an end-to-end transformer and graph neural network-based multi-camera tracking model. LaMMOn consists of three main modules: (1) Language Model Detection (LMD) for object detection; (2) Language and Graph Model Association module (LGMA) for object tracking and trajectory clustering; (3) Text-to-embedding module (T2E) that overcome the problem of data limitation by synthesizing the object embedding from defined texts. LaMMOn can be run online in real-time scenarios and achieve a competitive result on many datasets, e.g., CityFlow (HOTA 76.46%), I24 (HOTA 25.7%), and TrackCUIP (HOTA 80.94%) with an acceptable FPS (from 12.20 to 13.37) for an online application. [ABSTRACT FROM AUTHOR]

Objectives This article aims to enhance the performance of larger language models (LLMs) on the few-shot biomedical named entity recognition (NER) task by developing a simple and effective method called Retrieving and Chain-of-Thought (RT) framework and to evaluate the improvement after applying RT framework. Materials and Methods Given the remarkable advancements in retrieval-based language model and Chain-of-Thought across various natural language processing tasks, we propose a pioneering RT framework designed to amalgamate both approaches. The RT approach encompasses dedicated modules for information retrieval and Chain-of-Thought processes. In the retrieval module, RT discerns pertinent examples from demonstrations during instructional tuning for each input sentence. Subsequently, the Chain-of-Thought module employs a systematic reasoning process to identify entities. We conducted a comprehensive comparative analysis of our RT framework against 16 other models for few-shot NER tasks on BC5CDR and NCBI corpora. Additionally, we explored the impacts of negative samples, output formats, and missing data on performance. Results Our proposed RT framework outperforms other LMs for few-shot NER tasks with micro-F1 scores of 93.50 and 91.76 on BC5CDR and NCBI corpora, respectively. We found that using both positive and negative samples, Chain-of-Thought (vs Tree-of-Thought) performed better. Additionally, utilization of a partially annotated dataset has a marginal effect of the model performance. Discussion This is the first investigation to combine a retrieval-based LLM and Chain-of-Thought methodology to enhance the performance in biomedical few-shot NER. The retrieval-based LLM aids in retrieving the most relevant examples of the input sentence, offering crucial knowledge to predict the entity in the sentence. We also conducted a meticulous examination of our methodology, incorporating an ablation study. Conclusion The RT framework with LLM has demonstrated state-of-the-art performance on few-shot NER tasks. [ABSTRACT FROM AUTHOR]

To succeed in the information age, students need to learn to communicate their understanding of complex topics effectively. This is reflected in both educational standards and standardized tests. To improve their writing ability for highly structured domains like scientific explanations, students need feedback that accurately reflects the structure of their explanations — how they express causal relationships between concepts, and link them together. Unfortunately, providing that type of feedback is difficult and time-consuming for teachers. And Automated Essay Scoring (AES) systems generally aggregate the occurrence of features to provide holistic scores for essays, but give no information about the causal relationships identified by the students. In this paper, we evaluate a method for identifying causal relations in student scientific explanations based on recent advances in Natural Language Processing (NLP). Identification of the causal structure could be used to provide formative feedback to writers and their teachers about how the writing and reasoning could be improved. The identification method we evaluate here is based on Bidirectional Encoder Representations from Transformers (BERT, Devlin et al. , 2019), which uses unsupervised learning to create a general language model and then supervised learning to apply that model to a particular task. We compare a base BERT model to several of its special-purpose variants, and include two baseline approaches. The contributions of this study were that we determine which variants worked best along with their strengths and limitations for identifying causal structure in explanations. We found that the models detected the existence of a causal relationship with good performance, but identifying the particular causal relationship will require further research, largely due to the common challenge of imbalanced data classes. [ABSTRACT FROM AUTHOR]

Abstract The digitization of operation and maintenance in the intelligent power grid equipment relies on a diverse array of information for smart decision-making. In the domain of intelligent decision generation, proficiency is contingent upon extensive learning from copious amounts of text. This necessitates not only robust processing capabilities but also a high level of specialization. In addressing situations where authorization is lacking, pre-trained language models (PLMs) have already provided ideas when confronted with specialized domains or tasks. In consideration of the complexity of textual content in the field of the power grid, which encompasses a multitude of specialized knowledge and involves an abundance of proprietary terminology, we have undertaken an exploration of pre-trained model specialization using the power grid domain as an example, specifically for the task of generating maintenance strategies. A two-stage fine-tuning approach (P2FT) is employed, utilizing a large-scale pre-training model specifically designed for natural language processing. The efficacy and practical value of this method were evaluated through multiple metrics, juxtaposed with other advanced approaches involving low-parameter or parameter-free fine-tuning methods. Through a meticulous analysis and validation of experimental outcomes, we have corroborated the feasibility and practical application value of employing this approach for pre-trained model specialization. Additionally, it has furnished valuable guidance for text generation within both the Chinese language domain and the power grid domain.

Abstract Reports of Large Language Models (LLMs) passing board examinations have spurred medical enthusiasm for their clinical integration. Through a narrative review, we reflect upon the skill shifts necessary for clinicians to succeed in an LLM-enabled world, achieving benefits while minimizing risks. We suggest how medical education must evolve to prepare clinicians capable of navigating human-AI systems.

This study presents a novel approach to exploring the viewpoints of AI bots regarding the utilization of AI in business hotels using thematic analysis. Interviews were conducted with AI bots serving as the subjects. The rationale for this approach is that AI bots have undergone extensive training using a wide range of data from different internet sources. Therefore, having an interview with an AI bot can be viewed as an interaction with an entity that embodies collective viewpoints and information derived from the internet. The study has unearthed crucial themes that offer invaluable insights to industry participants and policymakers, contributing to the expansion of existing literature by generating novel knowledge and fostering a comprehensive understanding. The pioneering approach of incorporating AI bots as participants in interviews opens up fresh avenues for future research endeavours and engenders discussions and debates concerning the use of AI bots as participants.

Pre-trained language models (LMs) have shown remarkable success in diverse tasks and domains. An LM trained on the document of a specific area (e.g., biomedicine, education, and finance) provides expert-level knowledge about that domain, and there have been many efforts to develop such domain-specific LMs. Despite its potential benefits, however, developing LM in the technology commercialization (TC) domain has not been investigated. In this study, we build a TC-specialized large LM pre-trained on the Korean TC corpus. Firstly, we collect a large-scale dataset containing 199,857,586 general Korean sentences and 17,562,751 TC-related Korean sentences. Second, based on this large dataset, we pre-train a Transformer-based language model resulting in TC-BERT. Third, we validate TC-BERT on three practical applications: document classification, keyword extraction, and recommender system. For this, we devise a new keyword extraction algorithm and propose a document recommender algorithm based on TC-BERT’s document embedding. Through various quantitative and qualitative experiments, we comprehensively verify TC-BERT’s effectiveness and its application. [ABSTRACT FROM AUTHOR]

The study of bias in language models is a growing area of work, however, both research and resources are focused on English. In this paper, we make a first approach focusing on gender bias in some freely available Spanish language models trained using popular deep neural networks, like BERT or RoBERTa. Some of these models are known for achieving state-of-the-art results on downstream tasks. These promising results have promoted such models' integration in many real-world applications and production environments, which could be detrimental to people affected for those systems. This work proposes an evaluation framework to identify gender bias in masked language models, with explainability in mind to ease the interpretation of the evaluation results. We have evaluated 20 different models for Spanish, including some of the most popular pretrained ones in the research community. Our findings state that varying levels of gender bias are present across these models.This approach compares the adjectives proposed by the model for a set of templates. We classify the given adjectives into understandable categories and compute two new metrics from model predictions, one based on the internal state (probability) and the other one on the external state (rank). Those metrics are used to reveal biased models according to the given categories and quantify the degree of bias of the models under study. [ABSTRACT FROM AUTHOR]

The complex and rapidly evolving nature of modern software landscapes introduces challenges such as increasingly sophisticated cyber threats, the diversity in programming languages and coding styles, and the need to identify subtle patterns indicative of vulnerabilities. These hurdles underscore the necessity for advanced techniques that can effectively cope with the intricacies of software security. Hence, this paper gives a comparative empirical study in harnessing the potential of cutting-edge natural language processing (NLP) advancements, namely Word2Vec and CodeBERT to detect vulnerabilities in C and C++ programs in the proposed Defect-Scanner framework. With the capability of converting code components and source code into contextual embedding vectors, various potential NLP techniques are combined with several DL models to evaluate the precision and accuracy of identifying vulnerabilities within software systems. Moreover, the experimentations are conducted using datasets with different representation types of codes, aiming to figure out the best combination of NLP techniques and DL models to work with each form of input. As a result, besides the outperformance of CodeBERT-based models with accuracies of approximately 90%, this comparative study also provides a comprehensive evaluation of NLP-based software vulnerability detection in the face of intricate security challenges. [ABSTRACT FROM AUTHOR]

Technological advancements in web platforms allow people to express and share emotions toward textual write-ups written and shared by others. This brings about different interesting domains for analysis, emotion expressed by the writer and emotion elicited from the readers. In this paper, we propose a novel approach for readers' emotion detection from short-text documents using a deep learning model called REDAffectiveLM. Within state-of-the-art NLP tasks, it is well understood that utilizing context-specific representations from transformer-based pre-trained language models helps achieve improved performance. Within this affective computing task, we explore how incorporating affective information can further enhance performance. Toward this, we leverage context-specific and affect enriched representations by using a transformer-based pre-trained language model in tandem with affect enriched Bi-LSTM+Attention. For empirical evaluation, we procure a new dataset REN-20k, besides using RENh-4k and SemEval-2007. We evaluate the performance of our REDAffectiveLM rigorously across these datasets, against a vast set of state-of-the-art baselines, where our model consistently outperforms baselines and obtains statistically significant results. Our results establish that utilizing affect enriched representation along with context-specific representation within a neural architecture can considerably enhance readers' emotion detection. Since the impact of affect enrichment specifically in readers' emotion detection isn't well explored, we conduct a detailed analysis over affect enriched Bi-LSTM+Attention using qualitative and quantitative model behavior evaluation techniques. We observe that compared to conventional semantic embedding, affect enriched embedding increases the ability of the network to effectively identify and assign weightage to the key terms responsible for readers' emotion detection to improve prediction. [ABSTRACT FROM AUTHOR]

Abstract Natural language-based generative artificial intelligence (AI) has become increasingly prevalent in scientific research. Intriguingly, capabilities of generative pre-trained transformer (GPT) language models beyond the scope of natural language tasks have recently been identified. Here we explored how GPT-4 might be able to perform rudimentary structural biology modeling. We prompted GPT-4 to model 3D structures for the 20 standard amino acids and an α-helical polypeptide chain, with the latter incorporating Wolfram mathematical computation. We also used GPT-4 to perform structural interaction analysis between the anti-viral nirmatrelvir and its target, the SARS-CoV-2 main protease. Geometric parameters of the generated structures typically approximated close to experimental references. However, modeling was sporadically error-prone and molecular complexity was not well tolerated. Interaction analysis further revealed the ability of GPT-4 to identify specific amino acid residues involved in ligand binding along with corresponding bond distances. Despite current limitations, we show the current capacity of natural language generative AI to perform basic structural biology modeling and interaction analysis with atomic-scale accuracy.

As per the universal principles of usability and sustainable development goal defined for reducing inequality, each individual will have special abilities and challenges. Individuals with visual challenges have trouble interacting with digital platforms. In order to achieve inclusivity, there is a need to integrate universal accessibility into the web portals which are used as the most popular digital platforms. This study mainly focuses on the requirements of visually impaired users. The research work discusses the proposed approach for a telemedicine web platform by integrating voice navigation system. With this system, users can orally interact with the platform by using defined commands. Users can also receive the audio feedback from the computer for the specific command. The proposed speech recognition engine is implemented using deep learning models and tested on various browsers. The engine captures the commands in the form of user inputs and generates the proper audio feedback after executing the commands. Users with visual impairment have been involved in the evaluation by allowing them to interact on the telemedicine platform with verbal commands. The evaluation questions have been asked after each interaction to capture the response time, accuracy, experience and satisfaction. The outcome of the evaluation shows that individuals are showing significant progress in accessing required information and navigating the web pages on their choice of browsers. The study also observed that speech recognition engine along with speech grammar, acoustic model and synthesis has improved the usage of the system for all types of users. Ultimately, integrating voice navigation into web platforms provides a satisfactory experience to the users and improves accessibility, inclusivity and reduces inequality.

Abstract Accurate inference of orthologous genes constitutes a prerequisite for comparative and evolutionary genomics. SonicParanoid is one of the fastest tools for orthology inference; however, its scalability and accuracy have been hampered by time-consuming all-versus-all alignments and the existence of proteins with complex domain architectures. Here, we present a substantial update of SonicParanoid, where a gradient boosting predictor halves the execution time and a language model doubles the recall. Application to empirical large-scale and standardized benchmark datasets shows that SonicParanoid2 is much faster than comparable methods and also the most accurate. SonicParanoid2 is available at https://gitlab.com/salvo981/sonicparanoid2 and https://zenodo.org/doi/10.5281/zenodo.11371108 .

The article examines the use of artificial intelligence speech models for generating posts in social networks. Artificial intelligence (AI) is becoming increasingly integrated into our daily lives, and language models such as OpenAI's GPT-4 play a key role in this. Language models can be used for various tasks: automated translation, creating chatbots, and generating texts that match the style of a particular author or genre, formation personalized recommendations and even for detecting fake news. The article discusses how these models can be applied to generate content in social networks that is useful for personal blogs, commercial accounts, and news channels. Existing language models such as GPT, BERT, Gemini, and LLaMA were analyzed and their advantages and disadvantages were presented. The principle of operation of generative models and counting tokens when creating requests for artificial intelligence are considered, and relevant examples are given. The article describes the process of collecting and preparing data, filtering and optimizing it to reduce the cost of queries when training a model that provides the ability to generate content that corresponds to a given style and topic. The process of creating your own GPT application using the downloaded and processed dataset is described in detail. Comparative statistics of publications written by humans and artificial intelligence in the form of a table and graph are presented, and the results obtained are analyzed. In the future, the created GPT application will be able to integrate its own API services and use the DALL-E image generation model. This work highlights the importance and relevance of research in the field of AI language models and their application to generate content in social networks, opening up new opportunities for automating and optimizing these processes.

A study was conducted on the distribution data of a range of Russian national films from 2004 to September 2023 using machine learning methods, with successful and unsuccessful films and patriotic projects considered separately. The study utilized the ensemble machine learning model HistGradientBoostingClassifier and a sequential fully connected three-layer neural network based on the TensorFlow library, along with basic methods of natural language processing. It was found that patriotic films exhibit distribution characteristics lower than the market average and significantly lag behind successful Russian cinema projects at the box office. The study demonstrated the possibility of accurately predicting film box office receipts, distribution characteristics, as well as selecting project parameters and the composition of its creative team to enhance distribution results and increase the reach of target audiences. The study also illustrated the use of language models, specifically through film annotations, to create effective film content. A conclusion was drawn regarding the need for a set of measures to implement state policy in the field of education, information technology, culture, and art, focusing on the creation of unified media franchises and umbrella brands encompassing films, TV series, computer and online games, music, and other products. The implementation of the research results will enhance the effectiveness of public policy and the return on public and private- public investments in the creative industries sector, positioning cinema and the creative industry as self-sustaining and income-generating sectors of the economy. Leveraging rental data from other markets will facilitate the creation of content with higher demand globally, serving as a soft power and cultural influence tool. The results of the study can aid in project selection and film project development by the Russian Cinema Fund, the Ministry of Culture of the Russian Federation, private Russian investors and film studios, as well as authorities and organizations responsible for distributing Russian film content abroad. Проведено исследование массива прокатных данных российских национальных кинофильмов с февраля 2004 по сентябрь 2023 года с применением методов машинного обучения: отдельно рассмотрены успешные и неуспешные в прокате фильмы, проекты патриотической направленности. В исследовании использована ансамблевая модель машинного обучения HistGradientBoostingClassifier и последовательная полносвязная трехслойная нейросеть на основе библиотеки TensorFlow, базовые методы обработки естественного языка. Установлено, что патриотические фильмы имеют прокатные характеристики ниже, чем в среднем по рынку, и существенно отстают от успешных в прокате проектов российского кинематографа. Доказана возможность точного прогнозирования киносборов, прокатных характеристик, а также подбора параметров проекта и состава его творческой группы для улучшения результатов проката и увеличения охвата целевых аудиторий. Показано направление использования языковых моделей (на примере аннотаций кинофильмов) для создания эффективного киноконтента. Сделан вывод о необходимости комплекса мер по реализации государственной политики в сфере образования, информационных технологий, культуры и искусства, которая должна быть основана на создании единых медиафраншиз и «зонтичных брендов», включающих кинофильмы, сериалы, компьютерные, онлайн и настольные игры, музыкальную и иную продукцию. Внедрение результатов исследования повысит эффективность государственной политики, отдачу от государственных и частно-государственных инвестиций в сфере креативных индустрий. Кинематограф и креативная индустрия станут самоокупаемыми и приносящими доход секторами экономики. Использование данных проката других рынков позволит создавать более востребованный за рубежом контент как элемент «мягкой силы» и культурного влияния. Результаты исследования могут использоваться во время отбора и разработки кинопроектов «Фондом кино», Министерством культуры Российской Федерации, частными российскими инвесторами и киностудиями, органами власти и организациями, ответственными за распространение российского киноконтента за рубежом.

Natural language-based generative artificial intelligence (AI) has become increasingly prevalent in scientific research. Intriguingly, capabilities of generative pre-trained transformer (GPT) language models beyond the scope of natural language tasks have recently been identified. Here we explored how GPT-4 might be able to perform rudimentary structural biology modeling. We prompted GPT-4 to model 3D structures for the 20 standard amino acids and an α-helical polypeptide chain, with the latter incorporating Wolfram mathematical computation. We also used GPT-4 to perform structural interaction analysis between the anti-viral nirmatrelvir and its target, the SARS-CoV-2 main protease. Geometric parameters of the generated structures typically approximated close to experimental references. However, modeling was sporadically error-prone and molecular complexity was not well tolerated. Interaction analysis further revealed the ability of GPT-4 to identify specific amino acid residues involved in ligand binding along with corresponding bond distances. Despite current limitations, we show the current capacity of natural language generative AI to perform basic structural biology modeling and interaction analysis with atomic-scale accuracy. [ABSTRACT FROM AUTHOR]

Purpose: Clinical decision support systems (CDSS) are used to identify drugs with potential need for dose modification in patients with renal impairment. ChatGPT holds the potential to be integrated in the electronic health record (EHR) system to give such dosing advices. In this study, we aim to evaluate the performance of ChatGPT in clinical rule-guided dose interventions in hospitalized patients with renal impairment. Methods: This cross-sectional study was performed at Tergooi Medical Center, the Netherlands. CDSS alerts regarding renal dysfunction were collected from the electronic health record (EHR) during a 2-week period and were presented to ChatGPT and an expert panel. Alerts were presented with and without patient variables. To evaluate the performance, suggested medication interventions were compared. Results: In total, 172 CDDS alerts were generated for 80 patients. Indecisive responses by ChatGPT to alerts were excluded. For alerts presented without patient variables, ChatGPT provided "correct and identical" responses to 19.9%, "correct and different" responses to 26.7%, and "incorrect responses to 53.4% of the alerts. For alerts including patient variables, ChatGPT provided "correct and identical" responses to 16.7%, "correct and different" responses to 16.0%, and "incorrect responses to 67.3% of the alerts. Accuracy was better for newer drugs such as direct oral anticoagulants. Conclusion: The performance of ChatGPT in clinical rule-guided dose interventions in hospitalized patients with renal dysfunction was poor. Based on these results, we conclude that ChatGPT, in its current state, is not appropriate for automatic integration into our EHR to handle CDSS alerts related to renal dysfunction. [ABSTRACT FROM AUTHOR]

As per the universal principles of usability and sustainable development goal defined for reducing inequality, each individual will have special abilities and challenges. Individuals with visual challenges have trouble interacting with digital platforms. In order to achieve inclusivity, there is a need to integrate universal accessibility into the web portals which are used as the most popular digital platforms. This study mainly focuses on the requirements of visually impaired users. The research work discusses the proposed approach for a telemedicine web platform by integrating voice navigation system. With this system, users can orally interact with the platform by using defined commands. Users can also receive the audio feedback from the computer for the specific command. The proposed speech recognition engine is implemented using deep learning models and tested on various browsers. The engine captures the commands in the form of user inputs and generates the proper audio feedback after executing the commands. Users with visual impairment have been involved in the evaluation by allowing them to interact on the telemedicine platform with verbal commands. The evaluation questions have been asked after each interaction to capture the response time, accuracy, experience and satisfaction. The outcome of the evaluation shows that individuals are showing significant progress in accessing required information and navigating the web pages on their choice of browsers. The study also observed that speech recognition engine along with speech grammar, acoustic model and synthesis has improved the usage of the system for all types of users. Ultimately, integrating voice navigation into web platforms provides a satisfactory experience to the users and improves accessibility, inclusivity and reduces inequality. [ABSTRACT FROM AUTHOR]

Processing large and heterogeneous numbers of ideas submitted to crowdsourcing contests is a regular challenge for idea evaluators. The aim of this study is to investigate a potential use case for AI-based innovation management and to extend the knowledge of using automated novelty detection in idea evaluation processes. AI-based language models can automatically allocate short texts according to their semantic similarity in an embedded space. We represent the semantic content of crowdsourced ideas with the three contemporary text embeddings – Doc2Vec, SBERT, and GPT-3-based Ada Similarity – and compute their semantic distance to different reference sets using different novelty detection algorithms. We then compare the algorithm-generated scores with human novelty assessments to validate them. While selected novelty scores based on text embeddings correlate with humans, our results show that scores based on SBERT embeddings best match human novelty assessments. We also find that AI-based novelty detection approaches perform better for ideas below the median word count and when compared to a set of existing solutions, suggesting that the chosen language model is not the only factor influencing the applicability of the proposed approach. Furthermore, the study highlights important features and limitations of automatically generated novelty scores that need to be considered when complementing evaluators searching for new ideas in crowdsourcing contests and beyond. [ABSTRACT FROM AUTHOR]

Smooth interaction with a disaster‐affected community can create and strengthen its social capital, leading to greater effectiveness in the provision of successful post‐disaster recovery aid. To understand the relationship between the types of interaction, the strength of social capital generated, and the provision of successful post‐disaster recovery aid, intricate ethnographic qualitative research is required, but it is likely to remain illustrative because it is based, at least to some degree, on the researcher's intuition. This paper thus offers an innovative research method employing a quantitative artificial intelligence (AI)‐based language model, which allows researchers to re‐examine data, thereby validating the findings of the qualitative research, and to glean additional insights that might otherwise have been missed. This paper argues that well‐connected personnel and religiously‐based communal activities help to enhance social capital by bonding within a community and linking to outside agencies and that mixed methods, based on the AI‐based language model, effectively strengthen text‐based qualitative research. [ABSTRACT FROM AUTHOR]

Classifying domain entities into their respective top-level ontology concepts is a complex problem that typically demands manual analysis and deep expertise in the domain of interest and ontology engineering. Using an efficient approach to classify domain entities enhances data integration, interoperability, and the semantic clarity of ontologies, which are crucial for structured knowledge representation and modeling. Based on this, our main motivation is to help an ontology engineer with an automated approach to classify domain entities into top-level ontology concepts using informal definitions of these domain entities during the ontology development process. In this context, we hypothesize that the informal definitions encapsulate semantic information crucial for associating domain entities with specific top-level ontology concepts. Our approach leverages state-of-the-art language models to explore our hypothesis across multiple languages and informal definitions from different knowledge resources. In order to evaluate our proposal, we extracted multi-label datasets from the alignment of the OntoWordNet ontology and the BabelNet semantic network, covering the entire structure of the Dolce-Lite-Plus top-level ontology from most generic to most specific concepts. These datasets contain several different textual representation approaches of domain entities, including terms, example sentences, and informal definitions. Our experiments conducted 3 study cases, investigating the effectiveness of our proposal across different textual representation approaches, languages, and knowledge resources. We demonstrate that the best results are achieved using a classification pipeline with a K-Nearest Neighbor (KNN) method to classify the embedding representation of informal definitions from the Mistral large language model. The findings underscore the potential of informal definitions in reflecting top-level ontology concepts and point towards developing automated tools that could significantly aid ontology engineers during the ontology development process. [ABSTRACT FROM AUTHOR]

Smart contracts are decentralized applications that hold a pivotal role in blockchain-based systems. Smart contracts are composed of error-prone programming languages, so it is affected by many vulnerabilities (e.g., time dependence, outdated version, etc.), which can result in a substantial economic loss within the blockchain ecosystem. Therefore, many vulnerability detection tools are designed to detect the vulnerabilities in smart contracts such as Slither, Mythrill and so forth. However, these tools require high processing time and cannot achieve good accuracy with complex smart contracts nowadays. Consequently, many studies have shifted towards using Deep Learning (DL) techniques, which consider bytecode to determine vulnerabilities in smart contracts. However, these mechanisms reveal three main limitations. First, these mechanisms focus on multi-class problems, assuming that a given smart contract contains only a single vulnerability while the smart contract can contain more than one vulnerability. Second, these approaches encounter ineffective word embedding with large input sequences. Third, the learning model in these mechanisms is forced to classify into one of pre-defined labels even when it cannot make decisions accurately, leading to misclassifications. Therefore, in this paper, we propose a multi-label vulnerability classification mechanism using a language model. To deal with the ineffective word embedding, the proposed mechanism not only takes into account the implicit features derived from the language models (e.g., SecBERT, etc.) but also auxiliary features extracted from other word embedding techniques (e.g., TF-IDF, etc.). Besides, a trustworthy neural network model is proposed to reduce the misclassification rate of vulnerability classification. In detail, an additional neuron is added to the output of the model to indicate whether the model is able to make decisions accurately or not. The experimental results illustrate that the trustworthy model outperforms benchmarks (e.g., binary relevance, label powerset, classifier chain, etc.), achieving up to approximately 98% f1-score while requiring low execution time with 26 ms. [ABSTRACT FROM AUTHOR]

Purpose: The rise of artificial intelligence (AI) models like ChatGPT offers potential for varied applications, including patient education in healthcare. With gaps in osteoporosis and bone health knowledge and adherence to prevention and treatment, this study aims to evaluate the accuracy of ChatGPT in delivering evidence-based information related to osteoporosis. Methods: Twenty of the most common frequently asked questions (FAQs) related to osteoporosis were subcategorized into diagnosis, diagnostic method, risk factors, and treatment and prevention. These FAQs were sourced online and inputted into ChatGPT-3.5. Three orthopedic surgeons and one advanced practice provider who routinely treat patients with fragility fractures independently reviewed the ChatGPT-generated answers, grading them on a scale from 0 (harmful) to 4 (excellent). Mean response accuracy scores were calculated. To compare the variance of the means across the four categories, a one-way analysis of variance (ANOVA) was used. Results: ChatGPT displayed an overall mean accuracy score of 91%. Its responses were graded as "accurate requiring minimal clarification" or "excellent," with a mean response score ranging from 3.25 to 4. No answers were deemed inaccurate or harmful. No significant difference was observed in the means of responses across the defined categories. Conclusion: ChatGPT-3.5 provided high-quality educational content. It showcased a high degree of accuracy in addressing osteoporosis-related questions, aligning closely with expert opinions and current literature, with structured and inclusive answers. However, while AI models can enhance patient information accessibility, they should be used as an adjunct rather than a substitute for human expertise and clinical judgment. [ABSTRACT FROM AUTHOR]

In recent decades, antibodies have emerged as indispensable therapeutics for combating diseases, particularly viral infections. However, their development has been hindered by limited structural information and labor-intensive engineering processes. Fortunately, significant advancements in deep learning methods have facilitated the precise prediction of protein structure and function by leveraging co-evolution information from homologous proteins. Despite these advances, predicting the conformation of antibodies remains challenging due to their unique evolution and the high flexibility of their antigen-binding regions. Here, to address this challenge, we present the Bio-inspired Antibody Language Model (BALM). This model is trained on a vast dataset comprising 336 million 40% nonredundant unlabeled antibody sequences, capturing both unique and conserved properties specific to antibodies. Notably, BALM showcases exceptional performance across four antigen-binding prediction tasks. Moreover, we introduce BALMFold, an end-to-end method derived from BALM, capable of swiftly predicting full atomic antibody structures from individual sequences. Remarkably, BALMFold outperforms those well-established methods like AlphaFold2, IgFold, ESMFold and OmegaFold in the antibody benchmark, demonstrating significant potential to advance innovative engineering and streamline therapeutic antibody development by reducing the need for unnecessary trials. The BALMFold structure prediction server is freely available at https://beamlab-sh.com/models/BALMFold. [ABSTRACT FROM AUTHOR]

Large Language Models (LLMs) are artificial intelligence models that use deep neural networks to perform Natural Language Processing (NLP) tasks. These tasks include interaction between humans and computers, enabling computers to interpret and generate human languages in a meaningful manner. Large Language models are called "large" because of the architecture’s size and the huge sets of training text data. With the emergence of transformer-based LLMs, the game of NLPs has reached another level. This is due to their ability to handle long-range text dependencies in parallel. The growing prevalence of transformer-based LLMs in human lives has necessitated evaluating the scope of the Emotional Intelligence (EI) of LLMs. This paper will discuss the need for emotional intelligence in transformer-based LLMs and the various existing studies that have evaluated this aspect. The potential challenges of the LLMs along with the future directions for research in this field will also be discussed. [ABSTRACT FROM AUTHOR]

Multi-modal knowledge graphs (MMKG) integrate various modal information such as vision and text, presenting knowledge structures graphically. With the advancement of artificial intelligence, MMKG have played a significant role in recommendation systems, intelligent Q&A, and knowledge search among other fields. Compared to traditional knowledge graphs, MMKG can understand and present knowledge in multiple dimensions, possessing superior representation and application capabilities. To delve deep into the study of MMKG, this review first conducts a detailed analysis and elucidation of the value and categories of MMKG. Based on different construction methods, it compares and summarizes multimodal knowledge extraction, representation learning, entity alignment, and other aspects, categorizes multi-modal knowledge integration methods. It analyzes the progress in the applications of MMKG, discusses the limitations of MMKG, and proposes future research directions in the field of MMKG. [ABSTRACT FROM AUTHOR]

With the development of the world economy and the expansion of foreign trade, the status of English as a global language in English teaching is increasingly elevated, and language translation is an indispensable link. In the new intelligent translation and reading system based on language models, its translation part includes at least one dictionary used for translation. The system includes: reading and writing database, loading module, editing module, label module, trigger module, annotation module, and word addition module. Among them, the read-write database is used to store literature; The loading module is used to load article materials as reading documents. It adopts natural language processing technology to save system translation and reading time and improve system efficiency. In order to test the advantages of the system used in this article, 200 vocabulary words were input into the system and compared with the traditional system for translation accuracy. The translation accuracy of the system in this article was 94.64%, while the translation accuracy of the traditional system was 83.34%. The system in this article helps to improve translation efficiency and enhance reading experience. [ABSTRACT FROM AUTHOR]

Recent advancements in artificial intelligence (AI) present both opportunities and challenges within the scientific community. This study explores the capability of AI to replicate findings from genetic research, focusing on findings from prior work. Using an AI model without exposing any raw data, we created a dataset that closely mirrors the results of our original study, illustrating the ease of fabricating datasets with authenticity. This approach highlights the risks associated with AI misuse in scientific research. The study emphasizes the critical importance of maintaining the integrity of scientific inquiry in an era increasingly influenced by advanced AI technologies. [ABSTRACT FROM AUTHOR]

• We propose a language model-based method which uses differential encoding to generate dynamic word vectors for handling polysemous words in different datasets. • We introduce multi-task learning to share dynamic word vectors among different entity types, improving recognition and addressing data scarcity. • Our model excels over state-of-the-art methods in four standard training sets by boosting recall rates and enhancing accuracy through multi-task learning. Biomedical Named Entity Recognition (BioNER) is one of the most basic tasks in biomedical text mining, which aims to automatically identify and classify biomedical entities in text. Recently, deep learning-based methods have been applied to Biomedical Named Entity Recognition and have shown encouraging results. However, many biological entities are polysemous and ambiguous, which is one of the main obstacles to the task of biomedical named entity recognition. Deep learning methods require large amounts of training data, so the lack of data also affect the performance of model recognition. To solve the problem of polysemous words and insufficient data, for the task of biomedical named entity recognition, we propose a multi-task learning framework fused with language model based on the BiLSTM-CRF architecture. Our model uses a language model to design a differential encoding of the context, which could obtain dynamic word vectors to distinguish words in different datasets. Moreover, we use a multi-task learning method to collectively share the dynamic word vector of different types of entities to improve the recognition performance of each type of entity. Experimental results show that our model reduces the false positives caused by polysemous words through differentiated coding, and improves the performance of each subtask by sharing information between different entity data. Compared with other state-of-the art methods, our model achieved superior results in four typical training sets, and achieved the best results in F1 values. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Donghua University (English Edition) is the property of Journal of Donghua University Editorial Board and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)