General Language Research Articles

Da língua indígena à língua geral – algumas reflexões sobre a formação da Língua Geral Paulista

From indigenous language to general language – some reflections on the formation of the General Language of São Paulo: In Portuguese America, the general languages were the languages of indigenous origin used by the administration and the Church as an instrument of interethnic communication. These languages were spoken by everyone who was part of the colonial system. In the Brazilian territory, two general languageswere formed in very different linguistic contexts: the Língua Geral Paulista (LGP) and the Língua Geral Amazónica (LGA). Established in the 16th century, LGP emerged in the São Vicente region and was widely spread by the bandeirantes of São Paulo to other states in the 17th and 18th centuries. After a period of great expansion, and because of the political and social changes that took place in colonial society, this general language lost its hegemony in the 19th century and ceased to be spoken at the beginning of the 20th century. The article aims to analyse the socio-historical conditions that led to the formation, expansion, weakening and disappearance of the General Language of São Paulo.

DISCOURSE ANALYSIS OF PROPER NAMES OF UKRAINIAN EARLY MODERN PERIOD

specific character of proper names discursive implementa- of proper names discursive implementa- of proper names discursive implementa- proper names discursive implementa- discursive implementa- tion of Ukrainian proper names in the early modern period of the 17th century. The research task is to char- acterize proper names as a resource of persuasiveness in the preaching discursive practice of the 17th cen- tury; to determine the discursive polyplanarity of proper names in early modern Ukrainian sermons; to an- alyze proper names in terms of text generation and cognitive processes of the speaker (knowledge, mem- ory reproduction, analogy); to reveal the principles of combining proper names and creating images (met- aphors, similes) in Ukrainian early modern practice; to establish the influence of ideology on the discourse and creation of the imagery of agionyms for the designation of the names of the Rus saints. Research methods: discourse analysis (to characterize the process of text generation in the preach- ing discursive practice of the 17th century, which is conditioned by historical sociocultural processes), ono- mastic discourse analysis (based on the analysis of proper names in discourse) taking into account axiolog- ical and ideological representations of discourses, contextual and interpretive analysis (two stages, contex- tualization and interpretation, are covered), linguistic and cultural interpretation (to characterize baroque conceptism and determine secondary meanings in the preaching discursive practice). Proper names are an active resource of text creation in the early modern sermon, which is analyzed in accordance with the traditional episteme and sociocultural conditions of the Ukrainian early modern of the second half of the 17th century, the requirements of institutional discourse, the type of discursive practice, and the author’s instructions. Baroque conceptism with the creation of new meanings contributed to the memorization and perception of the presentation, activation of attention, emotions of recipients, influence on the consciousness of Ruthenianas. The preaching practice of this period acquires certain signs of epistemic discursive transformations, which explain beliefs and partly knowledge that acquire symbolic meaning in the preaching activity to the discourse of faith. The encyclopedic nature of the author, the creation of new images, and persuasiveness accumulate in the preaching of multi-layered proper names that synthesize different cultural layers in the preacher’s updated interpretation. The onomastic discourse analysis not only studies the imagery of historical discourses, but also extrapolates the study of proper names to the peculiarities of ancient text generation and the cognitive processes of the speaker. Processing of information by the author is considered in the aspects of processing foreign language sources (analysis of methods of adaptation of borrowed proper names), reproduction of information from memory (inaccurate spelling of proper names), creation of analogies, images (metaphors, similes). The creation of images by analogy determines baroque syncretism, which combines theonyms and mythonyms, biblical anthroponyms and mythonyms, Christian theonyms and pagan theonyms. New fields of knowledge and secondary meanings of astronomical names (names of constellations, stars) become one of the resources of the baroque imagery of the sermon, its axiological representation. Sociocultural and ideological conditions actualize sermons in honor of the Rus saints in the Ukrainian preaching discourse. The comparison of these agionyms with biblical anthroponyms (parallels from the Old Testament), names of ancient philosophers, poets (Sophocles), names of world history (Alexander the Great) is characterized.

Trustworthy AI: Securing Sensitive Data in Large Language Models

Large language models (LLMs) have transformed Natural Language Processing (NLP) by enabling robust text generation and understanding. However, their deployment in sensitive domains like healthcare, finance, and legal services raises critical concerns about privacy and data security. This paper proposes a comprehensive framework for embedding trust mechanisms into LLMs to dynamically control the disclosure of sensitive information. The framework integrates three core components: User Trust Profiling, Information Sensitivity Detection, and Adaptive Output Control. By leveraging techniques such as Role-Based Access Control (RBAC), Attribute-Based Access Control (ABAC), Named Entity Recognition (NER), contextual analysis, and privacy-preserving methods like differential privacy, the system ensures that sensitive information is disclosed appropriately based on the user’s trust level. By focusing on balancing data utility and privacy, the proposed solution offers a novel approach to securely deploying LLMs in high-risk environments. Future work will focus on testing this framework across various domains to evaluate its effectiveness in managing sensitive data while maintaining system efficiency.

The analysis of generative adversarial network in sports education based on deep learning

The importance of mental health is increasingly emphasized in modern society. The assessment of mental health qualities among college and university students as the future workforce holds significant significance. Therefore, this study, aiming to streamline the process of writing quality evaluations and enhance the fairness of assessment comments, explores the use of Generative Adversarial Network (GAN) technology in deep learning to evaluate the mental health qualities of college and university students through the unique avenue of sports. Firstly, GAN and Sequence Generative Adversarial Network (SeqGAN) models are introduced. Secondly, GAN is employed to construct a model for generating evaluation texts, encompassing the construction of a generator and discriminator, along with the introduction of a reward function. Finally, the constructed model is utilized to train on evaluation texts related to the mental health qualities of college and university students engaged in sports, validating the effectiveness of the model. The results indicate: (1) The pre-training of the generator in the constructed text generation model stabilizes after the 10th epoch. In contrast, the pre-training of the discriminator gradually stabilizes after the 35th epoch, demonstrating overall good training effectiveness. (2) When the generator’s update speed surpasses that of the discriminator, the model’s loss does not converge. However, with a reduction in the ratio of rounds between the two, there is a noticeable improvement in the convergence of the model. (3) The mean score of adaptability quality is the highest among the four indicators, suggesting a strong correlation between comment generation and adaptability quality. The results validate the effectiveness of the proposed text generation model in semantic control. This study aims to advance the level of mental health education among college and university students in the sports domain, providing theoretical references for enhancing the effectiveness of quality education assessments in other subjects as well.

A Comparative Analysis of Instruction Fine-Tuning Large Language Models for Financial Text Classification

Large Language Models (LLMs) have demonstrated impressive capabilities across diverse Natural Language Processing (NLP) tasks, including language understanding, reasoning, and generation. However, general-domain LLMs often struggle with financial tasks due to the technical and specialized nature of financial texts. This study investigates the efficacy of instruction fine-tuning smaller-scale LLMs, including Mistral-7B, Llama3-8B, and Phi3-mini, to enhance their performance in financial text classification tasks. We fine-tuned both instruction-tuned and base models across four financial classification tasks, achieving significant improvements in task-specific performance. Furthermore, we evaluated the zero-shot capabilities of these fine-tuned models on three unseen complex financial tasks, including argument classification, deal completeness classification, and causal classification. Our results indicate while base model fine-tuning led to greater degradation, instruction-tuned models maintained more robust performance. To address this degradation, we employed model merging techniques, integrating single-task domain-specific fine-tuned models with the base model. Using this merging method resulted in significant enhancements in zero-shot performance, even exceeding the original model’s accuracy on certain datasets. Our findings underscore the effectiveness of instruction fine-tuning and model merging for adapting LLMs to specialized financial text classification tasks.

The correlation between students' information literacy and the utilisation of AI text generator in completing academic studies

Background of the study: The use of AI Text Generator (ATG) provides many conveniences for students, especially in accessing information in the form of text to make research proposals as a requirement for completing academic studies in lectures, but on the other hand the use of this technology also has a negative impact if not used wisely. Purpose: To describe the correlation between student information literacy and the utilisation of AI Text Generator in completing academic studies Method: Simple correlation to determine the level or degree of correlation between two variables. Findings: There is a significant correlation between students' information literacy and its five indicators with the utilization of AI Text Generator in completing academic studies. The level of correlation is in the moderate category. The types of AI Text Generator most widely used by students are ChatGPT, Perplexity, and followed by several other types of ATG. The reasons students use ATGs include developing ideas and information, efficiency, writing assistance, concept development, reference, problem solving, and improving the quality of writing. Conclusion: The existence of a correlation with a level of correlation strength that is in the moderate category indicates that these two variables influence each other but are not completely dependent on each other.

Are large language models qualified reviewers in originality evaluation?

Large language models (LLMs) are a new generation of conversational language model with impressive semantic comprehension, text generation, and knowledge inference capabilities. LLMs are significantly influencing the development of science by assisting researchers in analyzing, understanding, and grasping original knowledge in scientific papers. This study investigates LLMs’ potential as qualified reviewers in originality evaluation in zero-shot learning, utilizing a unique, manually crafted prompt. Using biomedical papers as the data source, we constructed two evaluation datasets based on Nobel Prize papers and disruptive index. The evaluation performance of multiple LLMs of different types and scales on the datasets was scrutinized through the analysis of originality score (OS), originality type (OT), and originality description (OD), all of which were generated by the LLM. Our results show that LLMs can to some extent discern papers with distinct originality level via OS; however, they appear to be overly lenient reviewers. In LLMs’ evaluation mechanism, five distinct OTs reflecting varied research contributions do not manifest independently, but together they positively influence OS. Of all the LLMs analyzed, GPT-4 stood out as able to produce the most readable ODs, effectively explaining the inference process for both OS and OT from the perspectives of completeness, logicality, and regularity.

Interpretable Conversation Routing via the Latent Embeddings Approach

Large language models (LLMs) are quickly implemented to answer question and support systems to automate customer experience across all domains, including medical use cases. Models in such environments should solve multiple problems like general knowledge questions, queries to external sources, function calling and many others. Some cases might not even require a full-on text generation. They possibly need different prompts or even different models. All of it can be managed by a routing step. This paper focuses on interpretable few-shot approaches for conversation routing like latent embeddings retrieval. The work here presents a benchmark, a sorrow analysis, and a set of visualizations of the way latent embeddings routing works for long-context conversations in a multilingual, domain-specific environment. The results presented here show that the latent embeddings router is able to achieve performance on the same level as LLM-based routers with additional interpretability and higher level of control over model decision-making.

Exploring the effectiveness of instruction tuning in biomedical language processing

Large Language Models (LLMs), particularly those similar to ChatGPT, have significantly influenced the field of Natural Language Processing (NLP). While these models excel in general language tasks, their performance in domain-specific downstream tasks such as biomedical and clinical Named Entity Recognition (NER), Relation Extraction (RE), and Medical Natural Language Inference (NLI) is still evolving. In this context, our study investigates the potential of instruction tuning for biomedical language processing, applying this technique to two general LLMs of substantial scale. We present a comprehensive, instruction-based model trained on a dataset that consists of approximately 200,000 instruction-focused samples. This dataset represents a carefully curated compilation of existing data, meticulously adapted and reformatted to align with the specific requirements of our instruction-based tasks. This initiative represents an important step in utilising such models to achieve results on par with specialised encoder-only models like BioBERT and BioClinicalBERT for various classical biomedical NLP tasks. Our work includes an analysis of the dataset’s composition and its impact on model performance, providing insights into the intricacies of instruction tuning. By sharing our codes, models, and the distinctively assembled instruction-based dataset, we seek to encourage ongoing research and development in this area.22Our code repository is available at https://github.com/nlpie-research/BioInstTune-LLM.

An Exploration of Extractive method and Abstractive Method of Text Summarization with Various Approaches, Techniques and Datasets

Abstract: In today’s era, an enormous amount of data available which is in complex form such as Social media content, Images, Audio, Video and Text data form. The mechanism is very much needed to provide these types of data in simple and easy to understandable form. In this paper, importance is given on text summarization of large amount of textual data of documents must be specified in concise and understandable form. Automatic Text Summarization is the technique which is used to represent most significant concepts in precise and comprehensible form from original or source document to target document providing greater flexibility and convenience. Since the advancement in emerging technologies leads to the challenges in research field which is leading towards to need arises to analyze and should be resolved over the time period. Generally, there are two approaches for automatic text summarization: Extractive and Abstractive. The process of extractive based text summarization can be divided into two phases: pre-processing and processing. Automatic text summarization process model can be divided into three steps in the preprocessing step, source text interpretation to representation, source text representation transform to summary text representation with an algorithm. Finally, summary text generation from summary representation. In this paper discussion were made about some of the extractive based and abstractive based methods of text summarization for single documents. Various approaches of text summarization, methods needed for it along with tools and techniques, metrics needed to measure quality of generated text summary applications have been discussed in detail. This survey paper highlights the main advancement in the field of Automatic Text Summarization and provides the key methods with their benefits and limitations. Gives an overview of the past, present and future direction with certain solution to the challenges arises in text summary generation process with extractive and abstractive methods.

GENERAL AND LANGUAGE LEARNING SELF-EFFICACY BELIEFS OF GERMAN AND ENGLISH FOREIGN LANGUAGE LEARNERS

The process of learning a foreign language is complex and various internal and external factors have an impact on students’ academic success in the foreign language classroom. In the context of foreign language learning self-efficacy is defined as a person's perception of competence, rather than the actual level of competence, since confidence plays a fundamental role in foreign language learning.A student with a high level of self-efficacy tends to put more work into their studying, has higher levels of confidence and optimism when setbacks are encountered, and is more capable of addressing challenges, leading to an overall enhancement of their participation in learning.The present study aims to highlight the relationship between foreign language learning (German and English) and self-efficacy among undergraduate students of nursing, physiotherapy, midwifery, radiologic technology, and sanitary engineering of the Faculty of Health Studies, University of Mostar and establish if there are any differences in the way German and English learners perceive their general and foreign language self-efficacy. The study includes 53 undergraduate students studying at one of five majors and the Faculty of Health Studies in Mostar. Two scales were used in the study, the final version of the German Learning Self-Efficacy Belief Scale and the General Self-Efficacy Scale. The results reveal that the participants do not differ significantly on the levels for general and foreign language learning self-efficacy.Students of the Faculty of Health Studies have moderate levels of general and foreign language learning self-efficacy. These findings indicate that students regardless of the foreign language they study need to foster their abilities and perceptions in foreign language learning. Keywords: self efficacy, foreign language learning, German, English

LIPT: Improving Prompt Tuning with Late Inception Reparameterization

Prompt tuning is a mainstream technique for fine-tuning large language models (LLMs), offering minimal parameter adjustments by learning task-specific prompt vectors. However, it suffers from training costs due to network-wide backpropagation and weaker performance compared to methods like adapters and LoRA, likely due to the limited capacity of soft prompts to encode task-specific information. This study introduces Late Inception Prompt Tuning (LIPT), a novel approach to soft prompt learning that enhances performance and efficiency by shortening backpropagation paths and employing a multidimensional bottleneck network with greater capacity. LIPT surpasses existing prompt tuning techniques on various benchmark tasks, delivering a 1.3% gain over LPT and a 5% improvement compared to standard prompt tuning when applied to RoBERTa-large, while converging more rapidly. It achieves an average accuracy of 90% across ten benchmark datasets. Notably, in certain scenarios, LIPT’s performance approaches that of full-parameter fine-tuning methods. To evaluate parameter-efficient fine-tuning (PEFT) comprehensively, we propose an Efficiency Indicator (EI) that balances accuracy and cost. LIPT is well suited for natural language understanding tasks, like sentiment analysis and text classification, with potential extensions to larger-scale models and tasks like text generation. This framework advances the scalability and practicality of fine-tuning methods for diverse applications.

A general temperature-guided language model to design proteins of enhanced stability and activity.

Designing protein mutants with both high stability and activity is a critical yet challenging task in protein engineering. Here, we introduce PRIME, a deep learning model, which can suggest protein mutants with improved stability and activity without any prior experimental mutagenesis data for the specified protein. Leveraging temperature-aware language modeling, PRIME demonstrated superior predictive ability compared to current state-of-the-art models on the public mutagenesis dataset across 283 protein assays. Furthermore, we validated PRIME's predictions on five proteins, examining the impact of the top 30 to 45 single-site mutations on various protein properties, including thermal stability, antigen-antibody binding affinity, and the ability to polymerize nonnatural nucleic acid or resilience to extreme alkaline conditions. More than 30% of PRIME-recommended mutants exhibited superior performance compared to their premutation counterparts across all proteins and desired properties. We developed an efficient and effective method based on PRIME to rapidly obtain multisite mutants with enhanced activity and stability. Hence, PRIME demonstrates broad applicability in protein engineering.

Research on the Status Quo of Generative AI Tools Used by Pre-service Teachers

Through excellent text generation ability and rich knowledge reserves, generative AI has been widely discussed among pre-teachers in the past two years, and the analysis of the status of generative AI tools is particularly important. This study of 280 Chinese Qingdao University teachers questionnaire is analyzed, discusses the pre-professional teachers use the main scenarios, influencing factors of artificial intelligence, the view of former teachers use generative artificial intelligence tools on self-efficacy, and generative artificial intelligence tools for pre-service teachers cause certain anxiety. The following conclusions are drawn: generative artificial intelligence tools have improved the learning and teaching of teachers to a certain extent, examples are improving the efficiency of lesson preparation, enriching course design, summarizing knowledge points, etc. However, it still needs to be strengthened in the course feedback and personalized analysis. At the same time, it can be preliminarily concluded from this study that if people want to promote the further widespread application of generative AI tools in the pre-service group of teachers, the accuracy and some ethical considerations of generative AI tools need to be addressed. This study provides insights into the nexus between the characteristics of generative AI influencing pre-service teachers.

HealthWise: AI-Powered Medical Chatbot with Llama 2

This research paper presents the development of a medical chatbot utilizing cutting- edge technologies, specifically the Langchain framework and the Llama 2 language model, in conjunction with the PineCone database for efficient data storage and retrieval. The primary objective of this project is to create an intelligent conversational agent capable of providing reliable medical information, assisting users in symptom assessment, and guiding them towards appropriate healthcare resources. The integration of Langchain facilitates seamless interaction between the user interface and the underlying LLM, enabling natural language understanding and generation. Leveraging the advanced capabilities of the Llama 2 model ensures that the chatbot can engage in meaningful dialogues while maintaining context and relevance. Additionally, PineCone serves as a robust solution for managing and indexing medical data, enhancing the chatbot's ability to deliver accurate and context-aware responses. This study explores the design, implementation, and potential implications of our medical chatbot, contributing to the emerging field of AI-driven healthcare solutions and underscoring its role in improving patient engagement and access to medical advice. Keywords – Medical Chatbot, Langchain, Llama-2, Pine Cone, Natural Language Processing, Artificial Intelligence, Machine Learning, Generative Artificial Intelligence.

Advancements and Applications of Large Language Models in Natural Language Processing: A Comprehensive Review

Abstract. Large language models (LLMs) have revolutionized the field of natural language processing (NLP), demonstrating remarkable capabilities in understanding, generating, and manipulating human language. This comprehensive review explores the development, applications, optimizations, and challenges of LLMs. This paper begin by tracing the evolution of these models and their foundational architectures, such as the Transformer, GPT, and BERT. We then delve into the applications of LLMs in natural language understanding tasks, including sentiment analysis, named entity recognition, question answering, and text summarization, highlighting real-world use cases. Next, we examine the role of LLMs in natural language generation, covering areas such as content creation, language translation, personalized recommendations, and automated responses. We further discuss LLM applications in other NLP tasks like text style transfer, text correction, and language model pre-training. Subsequently, we explore techniques for optimizing and improving LLMs, including model compression, explainability, robustness, and security. Finally, we address the challenges posed by the significant computational requirements, sample inefficiency, and ethical considerations surrounding LLMs. We conclude by discussing potential future research directions, such as efficient architectures, few-shot learning, bias mitigation, and privacy-preserving techniques, which will shape the ongoing development and responsible deployment of LLMs in NLP.

ChatGPT: Technology Frontiers and Cybersecurity Challenges

This paper examines the evolution and application of OpenAI's advanced conversational AI, ChatGPT, particularly within the domain of cybersecurity. With an architecture built on the Transformer model, ChatGPT demonstrates significant capabilities in language understanding and generation. It leverages vast datasets, ranging from social media posts to technical documents, ensuring the model adapts to diverse fields and maintains compliance with privacy and security regulations. The paper explores ChatGPT's role in network security, highlighting its proficiency in threat detection, vulnerability assessment, and incident response, essential as regulations like GDPR and CCPA become more stringent. Furthermore, the study addresses potential security risks associated with AI, such as phishing and misinformation, and discusses mitigation strategies through advanced training techniques like adversarial training and multi-task learning. A novel variational autoencoder (VAE)-based method, T-VAE, is introduced, offering enhanced generalization capabilities across different tasks and scenarios. The findings suggest that while ChatGPT has made significant strides in cybersecurity applications, continuous improvements in model robustness and adaptability are necessary to mitigate emerging threats and adapt to evolving digital landscapes.

A Review of The Application of Natural Language Processing in Human-Computer Interaction

Over the past few decades, advancements in computing technologies have propelled research in Human-Computer Interaction (HCI). Recently, the rise of artificial intelligence, particularly deep learning and pre-trained models like BERT and GPT, has revolutionized Natural Language Processing (NLP), making HCI applications more intelligent and personalized. NLP has emerged as a critical technology in enhancing HCI, enabling more intuitive and efficient communication between users and machines. This review explores the various applications of NLP in HCI, highlighting its significant role in user interface design, chatbots, and virtual assistants. Specifically, the paper examines how NLP techniques such as intent recognition, sentiment analysis, and language generation contribute to the creation of more responsive and user-friendly interfaces through voice input, personalized experiences, and optimized feedback mechanisms. Furthermore, the challenges and limitations of implementing NLP in HCI are discussed, particularly concerning data privacy, model effectiveness, and the ethical implications of deploying NLP systems, with a focus on privacy and trustworthiness. Finally, the paper considers future research directions in this field, emphasizing the importance of interdisciplinary collaboration in overcoming current barriers and enhancing the applicability of NLP in real-world contexts, while urging developers to prioritize responsible AI practices in their designs.

Layer-Wise Learning Rate Optimization for Task-Dependent Fine-Tuning of Pre-Trained Models: An Evolutionary Approach

The superior performance of large-scale pre-trained models, such as Bidirectional Encoder Representations from Transformers (BERT) and Generative Pre-trained Transformer (GPT), has received increasing attention in both academic and industrial research and has become one of the current research hotspots. A pre-trained model refers to a model trained on large-scale unlabeled data, whose purpose is to learn general language representation or features for fine-tuning or transfer learning in subsequent tasks. After pre-training is complete, a small amount of labeled data can be used to fine-tune the model for a specific task or domain. This two-stage method of “pre-training+fine-tuning” has achieved advanced results in natural language processing (NLP) tasks. Despite widespread adoption, existing fixed fine-tuning schemes that adapt well to one NLP task may perform inconsistently on other NLP tasks given that different tasks have different latent semantic structures. In this article, we explore the effectiveness of automatic fine-tuning pattern search for layer-wise learning rates from an evolutionary optimization perspective. Our goal is to use evolutionary algorithms to search for better task-dependent fine-tuning patterns for specific NLP tasks than typical fixed fine-tuning patterns. Experimental results on two real-world language benchmarks and three advanced pre-training language models show the effectiveness and generality of the proposed framework.

Large language models in neurosurgery: a systematic review and meta-analysis.

Large Language Models (LLMs) have garnered increasing attention in neurosurgery and possess significant potential to improve the field. However, the breadth and performance of LLMs across diverse neurosurgical tasks have not been systematically examined, and LLMs come with their own challenges and unique terminology. We seek to identify key models, establish reporting guidelines for replicability, and highlight progress in key application areas of LLM use in the neurosurgical literature. We searched PubMed and Google Scholar using terms related to LLMs and neurosurgery ("large language model" OR "LLM" OR "ChatGPT" OR "GPT-3" OR "GPT3" OR "GPT-3.5" OR "GPT3.5" OR "GPT-4" OR "GPT4" OR "LLAMA" OR "MISTRAL" OR "BARD") AND "neurosurgery". The final set of articles was reviewed for publication year, application area, specific LLM(s) used, control/comparison groups used to evaluate LLM performance, whether the article reported specific LLM prompts, prompting strategy types used, whether the LLM query could be reproduced in its entirety (including both the prompt used and any adjoining data), measures of hallucination, and reported performance measures. Fifty-one articles met inclusion criteria, and were categorized into six application areas, with the most common being Generation of Text for Direct Clinical Use (n = 14, 27.5%), Answering Standardized Exam Questions (n = 12, 23.5%), and Clinical Judgement and Decision-Making Support (n = 11, 21.6%). The most frequently used LLMs were GPT-3.5 (n = 30, 58.8%), GPT-4 (n = 20, 39.2%), Bard (n = 9, 17.6%), and Bing (n = 6, 11.8%). Most studies (n = 43, 84.3%) used LLMs directly out-of-the-box, while 8 studies (15.7%) conducted advanced pre-training or fine-tuning. Large language models show advanced capabilities in complex tasks and hold potential to transform neurosurgery. However, research typically addresses basic applications and overlooks enhancing LLM performance, facing reproducibility issues. Standardizing detailed reporting, considering LLM stochasticity, and using advanced methods beyond basic validation are essential for progress.