Search

This report, developed as part of the 2024 UCSF Industry Documents Library Undergraduate Summer Fellowship, examines four distinct projects that leverage natural language processing and data science within the context of the JUUL Labs Collection and the broader IDL. Project One investigates the optical character recognition (OCR) accuracy of low-quality and handwritten documents in the absence of ground truth data. Project Two explores the implementation of embedding search algorithms and visualizations aimed at enhancing the relevance of document recommendations for users. Project Three employs txt-ferret to conduct a thorough scan of a substantial corpus of industry documents to identify sensitive information, including credit card numbers. Finally, Project Four assesses the biases present in large language model (LLM) summarization through the lens of sentiment analysis.

Purpose To assess the performance of a local open-source large language model (LLM) in various information extraction tasks from real-life emergency brain MRI reports. Materials and Methods All consecutive emergency brain MRI reports written in 2022 from a French quaternary center were retrospectively reviewed. Two radiologists identified MRI scans that were performed in the emergency department for headaches. Four radiologists scored the reports conclusions as either normal or abnormal. Abnormalities were labeled as either headache-causing or incidental. Vicuna (LMSYS Org), an open-source LLM, performed the same tasks. Vicunas performance metrics were evaluated using the radiologists consensus as the reference standard. Results Among the 2398 reports during the study period, radiologists identified 595 that included headaches in the indication (median age of patients, 35 years [IQR, 26-51 years]; 68% [403 of 595] women). A positive finding was reported in 227 of 595 (38%) cases, 136 of which could explain the headache. The LLM had a sensitivity of 98.0% (95% CI: 96.5, 99.0) and specificity of 99.3% (95% CI: 98.8, 99.7) for detecting the presence of headache in the clinical context, a sensitivity of 99.4% (95% CI: 98.3, 99.9) and specificity of 98.6% (95% CI: 92.2, 100.0) for the use of contrast medium injection, a sensitivity of 96.0% (95% CI: 92.5, 98.2) and specificity of 98.9% (95% CI: 97.2, 99.7) for study categorization as either normal or abnormal, and a sensitivity of 88.2% (95% CI: 81.6, 93.1) and specificity of 73% (95% CI: 62, 81) for causal inference between MRI findings and headache. Conclusion An open-source LLM was able to extract information from free-text radiology reports with excellent accuracy without requiring further training. Keywords: Large Language Model (LLM), Generative Pretrained Transformers (GPT), Open Source, Information Extraction, Report, Brain, MRI Supplemental material is available for this article. Published under a CC BY 4.0 license. See also the commentary by Akinci DAntonoli and Bluethgen in this issue.

Drug–drug interactions (DDIs) present a significant health burden, compounded by clinician time constraints and poor patient health literacy. We assessed the ability of ChatGPT (generative artificial intelligence‐based large language model) to predict DDIs in a real‐world setting. Demographics, diagnoses and prescribed medicines for 120 hospitalized patients were input through three standardized prompts to ChatGPT version 3.5 and compared against pharmacist DDI evaluation to estimate diagnostic accuracy. Area under receiver operating characteristic and inter‐rater reliability (Cohen's and Fleiss' kappa coefficients) were calculated. ChatGPT's responses differed based on prompt wording style, with higher sensitivity for prompts mentioning 'drug interaction'. Confusion matrices displayed low true positive and high true negative rates, and there was minimal agreement between ChatGPT and pharmacists (Cohen's kappa values 0.077–0.143). Low sensitivity values suggest a lack of success in identifying DDIs by ChatGPT, and further development is required before it can reliably assess potential DDIs in real‐world scenarios. [ABSTRACT FROM AUTHOR]

Artificial intelligence (AI) has become a vogue topic in the press, and descriptions of its potential impact range from apocalyptic to salvational. Interest in the topic will no doubt stimulate the search for applications to support both the technical and management systems aspects of process safety management. Within our industries, maintaining institutional memory and technical capability is made increasingly challenging by more frequent job movement among younger staff and the loss to the retirement of more senior staff. One would hope that AI could help fill the gaps caused by these factors. However, the author's sampling of current AI capabilities suggests that AI is not yet ready to do so. This paper provides some examples of errors and insufficiencies identified when seeking AI assistance in addressing process safety issues. It also suggests some existing challenges to better "training" of AI to support the needs of the process safety community. It concludes that caution should be applied, especially by less experienced personnel, when seeking AI assistance in addressing process safety–related technical matters. [ABSTRACT FROM AUTHOR]

The trend for an increasingly ubiquitous and cyber‐physical world has been leveraging the use and importance of microcontrollers (μC) to unprecedented levels. Therefore, microcontroller programming (μCP) becomes a paramount skill for electrical and computer engineering students. However, μCP poses significant challenges for undergraduate students, given the need to master low‐level programming languages and several algorithmic strategies that are not usual in "generic" programming. Moreover, μCP can be time‐consuming and complex even when using high‐level languages. This article samples the current state of μCP education in Portugal and unveils the potential support of natural language processing (NLP) tools (such as chatGPT). Our analysis of μCP curricular units from seven representative Portuguese engineering schools highlights a predominant use of AVR 8‐bit μC and project‐based learning. While NLP tools emerge as strong candidates as students' μC companion, their application and impact on the learning process and outcomes deserve to be understood. This study compares the most prominent NLP tools, analyzing their benefits and drawbacks for μCP education, building on both hands‐on tests and literature reviews. By providing automatic code generation and explanation of concepts, NLP tools can assist students in their learning process, allowing them to focus on software design and real‐world tasks that the μC is designed to handle, rather than on low‐level coding. We also analyzed the specific impact of chatGTP in the context of a μCP course at ISEP, confirming most of our expectations, but with a few curiosities. Overall, this work establishes the foundations for future research on the effective integration of NLP tools in μCP courses. [ABSTRACT FROM AUTHOR]

The development of intelligent medical service robots for patient care presents significant challenges, particularly in integrating diverse knowledge sources and enabling robots to autonomously perform tasks in dynamic and unpredictable healthcare environments. This study introduces a novel framework that combines large language models with healthcare-specific knowledge and robotic operations to enhance autonomous task execution for a Robotic Health Attendant. Utilizing OpenAI's ChatGPT, the system processes structured information about patient care protocols and unstructured human inputs to generate context-aware robot actions. A prototype system was tested in a simulated patient room where the robot successfully performed both simple individual actions and complex tasks involving the execution of multiple actions, based on real-time dialogues with the language model and predefined task specifications. The results demonstrate the potential of language models to reduce the reliance on hardcoded logic and provide healthcare professionals with the ability to interact with robotic systems through natural language. [ABSTRACT FROM AUTHOR]

Smart mining solutions currently suffer from inadequate big data support and insufficient AI applications. The main reason for these limitations is the absence of a comprehensive industrial internet cloud platform tailored for the coal industry, which restricts resource integration. This paper presents the development of an innovative platform designed to enhance safety, operational efficiency, and automation in fully mechanized coal mining in China. This platform integrates cloud edge computing, real-time data processing, and AI-driven analytics to improve decision-making and maintenance strategies. Several AI models have been developed for the proactive maintenance of comprehensive mining face equipment, including early warnings for periodic weighting and the detection of common faults such as those in the shearer, hydraulic support, and conveyor. The platform leverages large-scale knowledge graph models and Graph Retrieval-Augmented Generation (GraphRAG) technology to build structured knowledge graphs. This facilitates intelligent Q&A capabilities and precise fault diagnosis, thereby enhancing system responsiveness and improving the accuracy of fault resolution. The practical process of implementing such a platform primarily based on open-source components is summarized in this paper. [ABSTRACT FROM AUTHOR]

This article describes an experiment to use the ChatGPT Large Language Model as a tool to refine Schema.org metadata. ChatGPT was asked to give suggestions to improve a preexisting package of Schema.org structured metadata in the NMSU Library homepage for search engine optimization. A package of reformatted metadata based on ChatGPT's recommendations was used to replace the preexisting metadata for seven weeks and relevant web stats are compared to an equivalent seven-week period from the preceding semester. This article discusses ChatGPT's recommendations in some depth and examines the outcomes from a theoretical perspective. Implications of the experiment are outlined along with future areas for research. [ABSTRACT FROM AUTHOR]

Natural Language Processing (NLP) has shown promise in fields like radiology for converting unstructured into structured data, but acquiring suitable datasets poses several challenges, including privacy concerns. Specifically, we aim to utilize Large Language Models (LLMs) to extract medical information from dialogues between ambulance staff and patients to populate emergency protocol forms. However, we currently lack dialogues with known content that can serve as a gold standard for an evaluation. We designed a pipeline using the quantized LLM "Zephyr-7b-beta" for initial dialogue generation, followed by refinement and translation using OpenAI's GPT-4 Turbo. The MIMIC-IV database provided relevant medical data. The evaluation involved accuracy assessment via Retrieval-Augmented Generation (RAG) and sentiment analysis using multilingual models. Initial results showed a high accuracy of 94% with "Zephyr-7b-beta," slightly decreasing to 87% after refinement with GPT-4 Turbo. Sentiment analysis indicated a qualitative shift towards more positive sentiment post-refinement. These findings highlight the potential and challenges of using LLMs for generating synthetic medical dialogues, informing future NLP system development in healthcare. [ABSTRACT FROM AUTHOR]

This paper offers an in-depth review of the latest advancements in the automatic generation of medical case-based multiple-choice questions (MCQs). The automatic creation of educational materials, particularly MCQs, is pivotal in enhancing teaching effectiveness and student engagement in medical education. In this review, we explore various algorithms and techniques that have been developed for generating MCQs from medical case studies. Recent innovations in natural language processing (NLP) and machine learning (ML) for automatic language generation have garnered considerable attention. Our analysis evaluates and categorizes the leading approaches, highlighting their generation capabilities and practical applications. Additionally, this paper synthesizes the existing evidence, detailing the strengths, limitations, and gaps in current practices. By contributing to the broader conversation on how technology can support medical education, this review not only assesses the present state but also suggests future directions for improvement. We advocate for the development of more advanced and adaptable mechanisms to enhance the automatic generation of MCQs, thereby supporting more effective learning experiences in medical education. [ABSTRACT FROM AUTHOR]

Since the emergence of ChatGPT, research on large language models (LLMs) has actively progressed across various fields. LLMs, pre-trained on vast text datasets, have exhibited exceptional abilities in understanding natural language and planning tasks. These abilities of LLMs are promising in robotics. In general, traditional supervised learning-based robot intelligence systems have a significant lack of adaptability to dynamically changing environments. However, LLMs help a robot intelligence system to improve its generalization ability in dynamic and complex real-world environments. Indeed, findings from ongoing robotics studies indicate that LLMs can significantly improve robots' behavior planning and execution capabilities. Additionally, vision-language models (VLMs), trained on extensive visual and linguistic data for the vision question answering (VQA) problem, excel at integrating computer vision with natural language processing. VLMs can comprehend visual contexts and execute actions through natural language. They also provide descriptions of scenes in natural language. Several studies have explored the enhancement of robot intelligence using multimodal data, including object recognition and description by VLMs, along with the execution of language-driven commands integrated with visual information. This review paper thoroughly investigates how foundation models such as LLMs and VLMs have been employed to boost robot intelligence. For clarity, the research areas are categorized into five topics: reward design in reinforcement learning, low-level control, high-level planning, manipulation, and scene understanding. This review also summarizes studies that show how foundation models, such as the Eureka model for automating reward function design in reinforcement learning, RT-2 for integrating visual data, language, and robot actions in vision-language-action models, and AutoRT for generating feasible tasks and executing robot behavior policies via LLMs, have improved robot intelligence. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Frontiers of Computer Science & Technology is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. 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.)

As Natural Language Processing (NLP) continues to advance, driven by the emergence of sophisticated large language models such as ChatGPT, there has been a notable growth in research activity. This rapid uptake reflects increasing interest in the field and induces critical inquiries into ChatGPT's applicability in the NLP domain. This review paper systematically investigates the role of ChatGPT in diverse NLP tasks, including information extraction, Name Entity Recognition (NER), event extraction, relation extraction, Part of Speech (PoS) tagging, text classification, sentiment analysis, emotion recognition and text annotation. The novelty of this work lies in its comprehensive analysis of the existing literature, addressing a critical gap in understanding ChatGPT's adaptability, limitations, and optimal application. In this paper, we employed a systematic stepwise approach following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) framework to direct our search process and seek relevant studies. Our review reveals ChatGPT's significant potential in enhancing various NLP tasks. Its adaptability in information extraction tasks, sentiment analysis, and text classification showcases its ability to comprehend diverse contexts and extract meaningful details. Additionally, ChatGPT's flexibility in annotation tasks reduces manual efforts and accelerates the annotation process, making it a valuable asset in NLP development and research. Furthermore, GPT-4 and prompt engineering emerge as a complementary mechanism, empowering users to guide the model and enhance overall accuracy. Despite its promising potential, challenges persist. The performance of ChatGPT needs to be tested using more extensive datasets and diverse data structures. Subsequently, its limitations in handling domain-specific language and the need for fine-tuning in specific applications highlight the importance of further investigations to address these issues. [ABSTRACT FROM AUTHOR]

Recently, large language models (LLM) have made significant achievements in various fields. However, due to lack of specialized knowledge and the gap between modern medicine and traditional Chinese medicine (TCM), it is still a challenge to deploy LLM in TCM. Existing methods fail to maintain the structure of TCM prescription. To address the problems, a pattern of knowledge augmentation is proposed. The method includes model training, knowledge graph construction and knowledge augmentation. In the training phase, TCM language model is trained on TCM corpus, by a two-stage method combining pre-training and fine-tuning. In the knowledge graph construction phase, prescription knowledge graph is constructed from nearly 100000 preprocessed classical TCM prescriptions and those from ancient books. In the knowledge augmentation phase, enhanced by the above pattern, outputs are generated from computation of knowledge graph, according to the schema of knowledge graph from searching result, which preserves the structure of prescriptions. A set of evaluations specific to prescription optimizations is proposed, including objective and subjective indicators, to evaluate the performance of the model for the task. Experiment shows that the model improves greatly on both subjective and objective evaluations compared with baselines. BLEU-1 is increased by up to 0.09, while ROUGE-1 is increased by up to 0.21. Ablation study shows that, it is of vital importance for the model performance to be knowledge-augmented. BLEU-1 of augmentation-free model is decreased by about 37% compared with that of the augmented model.

The ever-growing number of pre-trained large language models (LLMs) across scientific domains presents a challenge for application developers. While these models offer vast potential, fine-tuning them with custom data, aligning them for specific tasks, and evaluating their performance remain crucial steps for effective utilization. However, applying these techniques to models with tens of billions of parameters can take days or even weeks on modern workstations, making the cumulative cost of model comparison and evaluation a significant barrier to LLM-based application development. To address this challenge, we introduce an end-to-end pipeline specifically designed for building conversational and programmable AI agents on high performance computing (HPC) platforms. Our comprehensive pipeline encompasses: model pre-training, fine-tuning, web and API service deployment, along with crucial evaluations for lexical coherence, semantic accuracy, hallucination detection, and privacy considerations. We demonstrate our pipeline through the development of chatHPC, a chatbot for HPC question answering and script generation. Leveraging our scalable pipeline, we achieve end-to-end LLM alignment in under an hour on the Frontier supercomputer. We propose a novel self-improved, self-instruction method for instruction set generation, investigate scaling and fine-tuning strategies, and conduct a systematic evaluation of model performance. The established practices within chatHPC will serve as a valuable guidance for future LLM-based application development on HPC platforms. [ABSTRACT FROM AUTHOR]

For a large number of citizens dealing with public authorities can feel like a harrowing experience. With a large amount of bureaucracy and ever-changing rules and procedures, it can be quite confusing for people to achieve their goals. The aim of this paper is to prove that a chatbot can be trained with minimal coding experience, by having a robust dataset created by people with domain-specific knowledge. This would empower every public institution to customise a helper chatbot. This chatbot would not only provide common information about required documents for whatever needs to be done, but also more general information pertaining to the specific aspects of that institution, such as its location, how to get to it, or public transportation connections nearby. The aim is to provide the necessary steps so that any public institution willing to undertake this digital transformation step can do so without much outside help or technical know-how in the field of Artificial Intelligence.

Relation extraction, as a key task in natural language processing, plays a significant role in deepening language understanding, constructing knowledge graphs, and optimizing information retrieval systems. However, traditional supervised learning methods are not well-suited for real-world scenarios due to the continuous emergence of new relations and the lack of large annotated datasets. Although the advent of large language models has significantly improved the performance of many natural language processing tasks, they still cannot effectively address the challenges of few-shot continual relation extraction. To fully leverage the semantic knowledge of large language models to mitigate catastrophic forgetting and overfitting issues, a novel few-shot continual relation extraction method, LAFA (large language model augmentation and feature alignment), is proposed. This method enhances representation alignment through various strategies such as relation instance rewriting, semantic expansion, and enhanced relation representation. It effectively improves the model adaptability to new relations and the retention of old knowledge while maintaining low data and computational costs. Experimental validation on two relation extraction datasets, FewRel and TACRED, demonstrates that LAFA outperforms existing methods in few-shot continual relation extraction tasks, particularly achieving the best results in incremental stages. Ablation experiments further reveal the significant contributions of each module to overall performance. Moreover, the inference efficiency and cost of LAFA are substantially lower than those of existing large language model-based methods, and it boasts strong scalability, being able to adapt to various language models.

Sentiment analysis is a vital task in natural language processing (NLP) that aims to identify and extract the emotional states and opinions of text. In this study, we conduct a comprehensive comparison of large language models (LLMs), such as ChatGPT and Google Bard, with conventional methods in sentiment analysis. We employ a rigorous evaluation framework that covers four essential metrics: accuracy, precision, recall, and the F1-score. Our results reveal that TextBlob outperforms other methods, achieving an impressive accuracy of 69% and precision of 83%. On the other hand, Bard shows a relatively poor performance, with only 39% accuracy and 46% precision. This study offers valuable insights into the diverse capabilities of AI models in sentiment analysis. A key finding of this study is the importance of model selection according to the specific requirements of the task. Each model has its own strengths and weaknesses, which are reflected in their performance profiles. Moreover, the context in which these models operate is crucial. For instance, ChatGPT generates varied responses, Bard struggles with multiple sentences, and Robustly Optimized BERT Pretraining Approach (RoBERTa) balances precision and recall. This study also reveals the performance gap between LLMs and state-of-the-art deep learning methods. We believe this work will inspire future research and applications of ChatGPT and similar AI models in sentiment analysis and related tasks. [ABSTRACT FROM AUTHOR]

Emerging research indicates that sentiment analyses of Dubrovnik focus mainly on hotel accommodations and restaurants. However, little attention has been paid to attractions, even though they are an important aspect of destinations and require more care and investment than amenities. This study examines how visitors experience Dubrovnik based on the reviews published on the Tripadvisor platform. Data were collected by implementing a web-scraping script to retrieve reviews of the tourist attraction "Old Town" from Tripadvisor, while data augmentation and random oversampling techniques were applied to address class imbalances. A sentiment analysis model, based on the pre-trained RoBERTa, was also developed and evaluated. In particular, a sentiment analysis was performed to compare reviews from 2022 and 2023. Overall, the results of this study are promising and demonstrate the effectiveness of this model and its potential applicability to other attractions. These findings provide valuable insights for decision makers to improve services and to increase visitor engagement. [ABSTRACT FROM AUTHOR]

Large language models (LLMs) are increasingly adopted in educational contexts to provide personalized support to students and teachers. The unprecedented capacity of LLM‐based applications to understand and generate natural language can potentially improve instructional effectiveness and learning outcomes, but the integration of LLMs in education technology has renewed concerns over algorithmic bias, which may exacerbate educational inequalities. Building on prior work that mapped the traditional machine learning life cycle, we provide a framework of the LLM life cycle from the initial development of LLMs to customizing pre‐trained models for various applications in educational settings. We explain each step in the LLM life cycle and identify potential sources of bias that may arise in the context of education. We discuss why current measures of bias from traditional machine learning fail to transfer to LLM‐generated text (eg, tutoring conversations) because text encodings are high‐dimensional, there can be multiple correct responses, and tailoring responses may be pedagogically desirable rather than unfair. The proposed framework clarifies the complex nature of bias in LLM applications and provides practical guidance for their evaluation to promote educational equity. Practitioner notesWhat is already known about this topic The life cycle of traditional machine learning (ML) applications which focus on predicting labels is well understood.Biases are known to enter in traditional ML applications at various points in the life cycle, and methods to measure and mitigate these biases have been developed and tested.Large language models (LLMs) and other forms of generative artificial intelligence (GenAI) are increasingly adopted in education technologies (EdTech), but current evaluation approaches are not specific to the domain of education.What this paper adds A holistic perspective of the LLM life cycle with domain‐specific examples in education to highlight opportunities and challenges for incorporating natural language understanding (NLU) and natural language generation (NLG) into EdTech.Potential sources of bias are identified in each step of the LLM life cycle and discussed in the context of education.A framework for understanding where to expect potential harms of LLMs for students, teachers, and other users of GenAI technology in education, which can guide approaches to bias measurement and mitigation.Implications for practice and/or policy Education practitioners and policymakers should be aware that biases can originate from a multitude of steps in the LLM life cycle, and the life cycle perspective offers them a heuristic for asking technology developers to explain each step to assess the risk of bias.Measuring the biases of systems that use LLMs in education is more complex than with traditional ML, in large part because the evaluation of natural language generation is highly context‐dependent (eg, what counts as good feedback on an assignment varies).EdTech developers can play an important role in collecting and curating datasets for the evaluation and benchmarking of LLM applications moving forward. [ABSTRACT FROM AUTHOR]

The emergence of large language models (LLMs) presents significant opportunities in accounting, including optimizing current processes, extracting new information, and updating accounting measurements. However, factors such as skill gaps, perceived complexity of integration, and cost constraints have limited the implementation of LLMs in accounting. This study provides an overview of mainstream LLM utilization methods, including user interface and application programming interface, and introduces a novel approach via robotic process automation integration. The advantages and limitations of each method are discussed, accompanied by a current analysis of the time, labor, and monetary costs involved in using LLMs for accounting tasks. To facilitate practical applications, three off-the-shelf examples are provided. This study contributes to the literature and practice by summarizing and comparing LLM implementation methods, responding to the challenges raised by researchers and stakeholders, and bridging the gap between technology innovation and its practical application in accounting. JEL Classifications: M41; O14; O33; D61. [ABSTRACT FROM AUTHOR]

This study introduces a novel approach to personalized information retrieval by integrating retrieval augmentation generation (RAG) with a personalized database system. Recent advancements in large language models (LLMs) have shown impressive text generation capabilities but face limitations in knowledge accuracy and hallucinations. Our research addresses these challenges by combining LLMs with structured, personalized data to enhance search precision and relevance. By tagging keywords within personal documents and organizing information into context-based categories, users can conduct efficient searches within their data repositories. We conducted experiments using the GPT-3.5 and text-embedding-ada-002 models and evaluated the RAG assessment framework with five different language models and two embedding models. Our results indicate that the combination of GPT-3.5 and text-embedding-ada-002 is effective for a personalized database question-answering system, with potential for various language models depending on the application. Our approach offers improved accuracy, real-time data updates, and enhanced user experience, making a significant contribution to information retrieval by LLMs and impacting various artificial intelligence applications. [ABSTRACT FROM AUTHOR]

It seems, in many quarters, that technology is viewed as bad for aspirations for a sustainable future. This Neo‐Luddite narrative is hampering the way the advantages and disadvantages of digital technology are perceived. Mario Carpo, Reyner Banham Professor of Architectural History and Theory at the Bartlett School of Architecture, University College London (UCL), argues that new technologies succeed older ones because they perform tasks better and more efficiently, innovatively and faster, both changing and responding to changes in societal and urban realms. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Frontiers of Computer Science & Technology is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. 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.)

Unbelievable progress in creating real-looking texts, audios and videos, numerous product designs and templates using generative artificial intelligence (GenAI)-enabled technologies have made work easier. However, they have their possible ill-effects too. Since the fruition of regulatory framework trails technological evolution, the future can see a virus-antivirus kind of ecosystem, where one set of GenAI tools will create contents that look most authentic and evade plagiarism and the other set will be busy trapping those. This study delves deep into the working of GenAI, various industry use cases, risk paradigms, and how organizations can start adopting the same, as it may be impossible to deny GenAI's business benefits for long. [ABSTRACT FROM AUTHOR]

A good theory of mathematical beauty is more practical than any current observation, as new predictions about physical reality can be self-consistently verified. This belief applies to the current status of understanding deep neural networks including large language models and even the biological intelligence. Toy models provide a metaphor of physical reality, allowing mathematically formulating the reality (i.e., the so-called theory), which can be updated as more conjectures are justified or refuted. One does not need to present all details in a model, but rather, more abstract models are constructed, as complex systems such as the brains or deep networks have many sloppy dimensions but much less sti dimensions that strongly impact macroscopic observables. This type of bottom-up mechanistic modeling is still promising in the modern era of understanding the natural or artificial intelligence. Here, we shed light on eight challenges in developing theory of intelligence following this theoretical paradigm. Theses challenges are representation learning, generalization, adversarial robustness, continual learning, causal learning, internal model of the brain, next-token prediction, and the mechanics of subjective experience. [ABSTRACT FROM AUTHOR]

With the development of deep learning, a higher level of perception of the environment such as the semantic level can be achieved in the simultaneous localization and mapping (SLAM) domain. However, previous works did not achieve a natural-language level of perception. Therefore, LP-SLAM (Language-Perceptive RGB-D SLAM) is proposed that leverages large language models (LLMs). The texts in the scene can be detected by scene text recognition (STR) and mapped as landmarks with a task-driven selection. A text error correction chain (TECC) is designed with a similarity classification method, a two-stage memory strategy, and a text clustering method. The proposed architecture is designed to deal with the mis-detection and mis-recognition cases of STR and to provide accurate text information to the framework. The proposed framework takes input images and generates a 3D map with sparse point cloud and task-related texts. Finally, a natural user interface (NUI) is designed based on the constructed map and LLM, which gives position instructions based on users' natural queries. The experimental results validated the proposed TECC design and the overall framework. We publish the virtual dataset with ground truth, as well as the source code for further research. https://github.com/GroupOfLPSLAM/LP_SLAM. [ABSTRACT FROM AUTHOR]

Purpose: To assess the performance of GPT-4 Turbo with Vision (GPT-4TV), OpenAI's latest multimodal large language model, by comparing its ability to process both text and image inputs with that of the text-only GPT-4 Turbo (GPT-4 T) in the context of the Japan Diagnostic Radiology Board Examination (JDRBE). Materials and methods: The dataset comprised questions from JDRBE 2021 and 2023. A total of six board-certified diagnostic radiologists discussed the questions and provided ground-truth answers by consulting relevant literature as necessary. The following questions were excluded: those lacking associated images, those with no unanimous agreement on answers, and those including images rejected by the OpenAI application programming interface. The inputs for GPT-4TV included both text and images, whereas those for GPT-4 T were entirely text. Both models were deployed on the dataset, and their performance was compared using McNemar's exact test. The radiological credibility of the responses was assessed by two diagnostic radiologists through the assignment of legitimacy scores on a five-point Likert scale. These scores were subsequently used to compare model performance using Wilcoxon's signed-rank test. Results: The dataset comprised 139 questions. GPT-4TV correctly answered 62 questions (45%), whereas GPT-4 T correctly answered 57 questions (41%). A statistical analysis found no significant performance difference between the two models (P = 0.44). The GPT-4TV responses received significantly lower legitimacy scores from both radiologists than the GPT-4 T responses. Conclusion: No significant enhancement in accuracy was observed when using GPT-4TV with image input compared with that of using text-only GPT-4 T for JDRBE questions. [ABSTRACT FROM AUTHOR]

Background and Objectives: Integrating large language models (LLMs) such as GPT-4 Turbo into diagnostic imaging faces a significant challenge, with current misdiagnosis rates ranging from 30–50%. This study evaluates how prompt engineering and confidence thresholds can improve diagnostic accuracy in neuroradiology. Methods: We analyze 751 neuroradiology cases from the American Journal of Neuroradiology using GPT-4 Turbo with customized prompts to improve diagnostic precision. Results: Initially, GPT-4 Turbo achieved a baseline diagnostic accuracy of 55.1%. By reformatting responses to list five diagnostic candidates and applying a 90% confidence threshold, the highest precision of the diagnosis increased to 72.9%, with the candidate list providing the correct diagnosis at 85.9%, reducing the misdiagnosis rate to 14.1%. However, this threshold reduced the number of cases that responded. Conclusions: Strategic prompt engineering and high confidence thresholds significantly reduce misdiagnoses and improve the precision of the LLM diagnostic in neuroradiology. More research is needed to optimize these approaches for broader clinical implementation, balancing accuracy and utility. [ABSTRACT FROM AUTHOR]

The study aims to identify the knowledge, skills and competencies required by accounting and auditing (AA) professionals in the context of integrating disruptive Generative Artificial Intelligence (GenAI) technologies and to develop a framework for integrating GenAI capabilities into organisational systems, harnessing its potential to revolutionise lifelong learning and skills development and to assist day-to-day operations and decision-making. Through a systematic literature review, 103 papers were analysed, to outline, in the current business ecosystem, the competencies' demand generated by AI adoption and, in particular, GenAI and its associated risks, thus contributing to the body of knowledge in underexplored research areas. Positioned at the confluence of accounting, auditing and GenAI, the paper introduces a meaningful overview of knowledge in the areas of effective data analysis, interpretation of findings, risk awareness and risk management. It emphasizes and reshapes the role of required skills for accounting and auditing professionals in discovering the true potential of GenAI and adopting it accordingly. The study introduces a new LLM-based system model that can enhance its GenAI capabilities through collaboration with similar systems and provides an explanatory scenario to illustrate its applicability in the accounting and audit area. [ABSTRACT FROM AUTHOR]

Copyright of Romanian Journal of Information Technology & Automatic Control / Revista Română de Informatică și Automatică is the property of National Institute for Research & Development in Informatics - ICI Bucharest 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.)

The major objective of the present research is to inspect sentiment analysis models that have been trained on Twitter corpus by utilising the Large Language Model (LLM) gpt-3–5-turbo-16k version of the Generative Pretrained Transformer (GPT 3.5) model. Such trained models include the Bidirectional long short-term memory neural network (BiLSTM), Convolutional Neural Networks, Gated Recurrent Unit and Recurrent Neural Network which were used to perform computational tasks on the Ohio Supercomputer and Utkarsh Server, in comparison to work conducted on Google Colab and a Personal Computer (PC).This research also looks at the performance as well as the computational aspects of these models in terms of accuracy, recall, F1-score, time/memory complexity and resource requirements (CPU/GPU) throughout the training and testing phases of each model. The training accuracies in this case were concentrated between 49.91 % and 99.98 % while those of the testing accuracies of the models accounts for about 50.00–75.00 %. For instance, models including BiLSTM and RNN usually exhibit more time complexity because of the nature of the models (sequential computation), on the contrary, CNNs are less time-consuming and are more effective in terms of storage modifying layered architecture. The use of supercomputers and specialized servers reduces training time, but resource constraints on platforms such as personal computers or Colab cause considerable divergence.

The Design Structure Matrix (DSM) is an established method used in dependency modelling, especially in the design of complex engineering systems. The generation of DSM is traditionally carried out through manual means and can involve interviewing experts to elicit critical system elements and the relationships between them. Such manual approaches can be time-consuming and costly. This paper presents a workflow that uses a Large Language Model (LLM) to support the generation of DSM and improve productivity. A prototype of the workflow was developed in this work and applied on a diesel engine DSM published previously. It was found that the prototype could reproduce 357 out of 462 DSM entries published (i.e. 77.3%), suggesting that the work can aid DSM generation. A no-code version of the prototype is made available online to support future research.

In recent years, large language models (LLMs) have seen rapid advancement and adoption, and are increasingly being used in educational contexts. In this perspective article, we explore the open challenge of leveraging LLMs to create personalized learning environments that support the “whole learner” by modeling and adapting to both cognitive and non-cognitive characteristics. We identify three key challenges toward this vision: (1) improving the interpretability of LLMs' representations of whole learners, (2) implementing adaptive technologies that can leverage such representations to provide tailored pedagogical support, and (3) authoring and evaluating LLM-based educational agents. For interpretability, we discuss approaches for explaining LLM behaviors in terms of their internal representations of learners; for adaptation, we examine how LLMs can be used to provide context-aware feedback and scaffold non-cognitive skills through natural language interactions; and for authoring, we highlight the opportunities and challenges involved in using natural language instructions to specify behaviors of educational agents. Addressing these challenges will enable personalized AI tutors that can enhance learning by accounting for each student's unique background, abilities, motivations, and socioemotional needs.