Search

The object of the research is the task of automatic abstracting of natural language texts. The importance of these tasks is determined by the existing problem of creating essays that would adequately reflect the content of the original text and highlight key information. This task requires the ability of models to deeply analyze the context of the text, which complicates the abstracting process. Results are presented that demonstrate the effectiveness of using generative models based on neural networks, text semantic analysis methods, and deep learning for automatic creation of abstracts. The use of models showed a high level of adequacy and informativeness of abstracts. GPT (Generative Pre-trained Transformer) generates text that looks like it was written by a human, which makes it useful for automatic essay generation. For example, the GPT model generates abbreviated summaries based on input text, while the BERT model is used for summarizing texts in many areas, including search engines and natural language processing. This allows for short but informative abstracts that retain the essential content of the original and provides the ability to produce high-quality abstracts that can be used for abstracting web pages, emails, social media, and other content. Compared to traditional abstracting methods, artificial intelligence provides such advantages as greater accuracy, informativeness and the ability to process large volumes of text more efficiently, which facilitates access to information and improves productivity in text processing. Automatic abstracting of texts using artificial intelligence models allows to significantly reduce the time required for the analysis of large volumes of textual information. This is especially important in today’s information environment, where the amount of available data is constantly growing. The use of these models promotes efficient use of resources and increases overall productivity in a variety of fields, including scientific research, education, business and media [ABSTRACT FROM AUTHOR]

Motivation Citations have a fundamental role in scholarly communication and assessment. Citation accuracy and transparency is crucial for the integrity of scientific evidence. In this work, we focus on quotation errors, errors in citation content that can distort the scientific evidence and that are hard to detect for humans. We construct a corpus and propose natural language processing (NLP) methods to identify such errors in biomedical publications. Results We manually annotated 100 highly-cited biomedical publications (reference articles) and citations to them. The annotation involved labeling citation context in the citing article, relevant evidence sentences in the reference article, and the accuracy of the citation. A total of 3063 citation instances were annotated (39.18% with accuracy errors). For NLP, we combined a sentence retriever with a fine-tuned claim verification model to label citations as ACCURATE, NOT_ACCURATE, or IRRELEVANT. We also explored few-shot in-context learning with generative large language models. The best performing model—which uses citation sentences as citation context, the BM25 model with MonoT5 reranker for retrieving top-20 sentences, and a fine-tuned MultiVerS model for accuracy label classification—yielded 0.59 micro-F1 and 0.52 macro-F1 score. GPT-4 in-context learning performed better in identifying accurate citations, but it lagged for erroneous citations (0.65 micro-F1, 0.45 macro-F1). Citation quotation errors are often subtle, and it is currently challenging for NLP models to identify erroneous citations. With further improvements, the models could serve to improve citation quality and accuracy. Availability and implementation We make the corpus and the best-performing NLP model publicly available at https://github.com/ScienceNLP-Lab/Citation-Integrity/. [ABSTRACT FROM AUTHOR]

Artificial intelligence (AI) has become increasingly used in neurosurgery. Generative pretrained transformers (GPTs) have been of particular interest. However, ethical concerns regarding the incorporation of AI into the field remain underexplored. We delineate key ethical considerations using a novel GPT-based, human-modified approach, synthesize the most common considerations, and present an ethical framework for the involvement of AI in neurosurgery. GPT-4, ChatGPT, Bing Chat/Copilot, You, Perplexity.ai, and Google Bard were queried with the prompt "How can artificial intelligence be ethically incorporated into neurosurgery?". Then, a layered GPT-based thematic analysis was performed. The authors synthesized the results into considerations for the ethical incorporation of AI into neurosurgery. Separate Pareto analyses with 20% threshold and 10% threshold were conducted to determine salient themes. The authors refined these salient themes. Twelve key ethical considerations focusing on stakeholders, clinical implementation, and governance were identified. Refinement of the Pareto analysis of the top 20% most salient themes in the aggregated GPT outputs yielded 10 key considerations. Additionally, from the top 10% most salient themes, 5 considerations were retrieved. An ethical framework for the use of AI in neurosurgery was developed. It is critical to address the ethical considerations associated with the use of AI in neurosurgery. The framework described in this manuscript may facilitate the integration of AI into neurosurgery, benefitting both patients and neurosurgeons alike. We urge neurosurgeons to use AI only for validated purposes and caution against automatic adoption of its outputs without neurosurgeon interpretation. [ABSTRACT FROM AUTHOR]

The article discusses the advancements and trends in oral and maxillofacial surgery. It highlights the integration of cutting-edge technology, personalized treatment modalities, and minimally invasive procedures in the field. The use of digital imaging, 3D modeling, and AI is transforming the way surgeries are planned and executed, leading to improved outcomes and patient satisfaction. The article also explores specific studies on topics such as challenging airway situations, total joint replacement for reconstruction, fat grafting in parotidectomy surgeries, and the impact of surgical precision in maxillomandibular advancement. The collection of articles demonstrates the progress and potential of oral and maxillofacial surgery in providing innovative and patient-centered care. [Extracted from the article]

Background and Aims: Artificial intelligence (AI) chatbots like Conversational Generative Pre-trained Transformer (ChatGPT) have recently created much buzz, especially regarding patient education. Such informed patients understand and adhere to the management and get involved in shared decision making. The accuracy and understandability of the generated educational material are prime concerns. Thus, we compared ChatGPT with traditional patient information leaflets (PILs) about chronic pain medications. Methods: Patients' frequently asked questions were generated from PILs available on the official websites of the British Pain Society (BPS) and the Faculty of Pain Medicine. Eight blinded annexures were prepared for evaluation, consisting of traditional PILs from the BPS and AI-generated patient information materials structured similar to PILs by ChatGPT. The authors performed a comparative analysis to assess materials' readability, emotional tone, accuracy, actionability, and understandability. Readability was measured using Flesch Reading Ease (FRE), Gunning Fog Index (GFI), and Flesch-Kincaid Grade Level (FKGL). Sentiment analysis determined emotional tone. An expert panel evaluated accuracy and completeness. Actionability and understandability were assessed with the Patient Education Materials Assessment Tool. Results: Traditional PILs generally exhibited higher readability (P values < 0.05), with [mean (standard deviation)] FRE [62.25 (1.6) versus 48 (3.7)], GFI [11.85 (0.9) versus 13.65 (0.7)], and FKGL [8.33 (0.5) versus 10.23 (0.5)] but varied emotional tones, often negative, compared to more positive sentiments in ChatGPT-generated texts. Accuracy and completeness did not significantly differ between the two. Actionability and understandability scores were comparable. Conclusion: While AI chatbots offer efficient information delivery, ensuring accuracy and readability, patient-centeredness remains crucial. It is imperative to balance innovation with evidence-based practice. [ABSTRACT FROM AUTHOR]

Chat Generative Pretrained Transformer (ChatGPT) is a large language model capable of generating human-like text. This study sought to evaluate ChatGPT's performance on Surgical Council on Resident Education (SCORE) self-assessment questions. General surgery multiple choice questions were randomly selected from the SCORE question bank. ChatGPT (GPT-3.5, April-May 2023) evaluated questions and responses were recorded. ChatGPT correctly answered 123 of 200 questions (62%). ChatGPT scored lowest on biliary (2/8 questions correct, 25%), surgical critical care (3/10, 30%), general abdomen (1/3, 33%), and pancreas (1/3, 33%) topics. ChatGPT scored higher on biostatistics (4/4 correct, 100%), fluid/electrolytes/acid-base (4/4, 100%), and small intestine (8/9, 89%) questions. ChatGPT answered questions with thorough and structured support for its answers. It scored 56% on ethics questions and provided coherent explanations regarding end-of-life discussions, communication with coworkers and patients, and informed consent. For many questions answered incorrectly, ChatGPT provided cogent, yet factually incorrect descriptions, including anatomy and steps of operations. In two instances, it gave a correct explanation but chose the wrong answer. It did not answer two questions, stating it needed additional information to determine the next best step in treatment. ChatGPT answered 62% of SCORE questions correctly. It performed better at questions requiring standard recall but struggled with higher-level questions that required complex clinical decision making, despite providing detailed responses behind its rationale. Due to its mediocre performance on this question set and sometimes confidently-worded, yet factually inaccurate responses, caution should be used when interpreting ChatGPT's answers to general surgery questions. • ChatGPT scored 62% on general surgery in-service training exam preparation questions. • ChatGPT performed better at questions regarding rote memorization, but struggled with questions requiring more complex clinical decision-making. • Given limited accuracy and incorrect answers, caution should be used when interpreting ChatGPT's answers regarding general surgery topics. [ABSTRACT FROM AUTHOR]

This article reviews the existing literature on the use of conversational AI robots based on GPT language models in libraries, focusing on their viability, usefulness and performance in answering reference questions. The main objectives are to synthesize current research, evaluate the performance of GPT models and reflect on librarians' positioning regarding the integration of these technologies into library reference services. The methodology includes the analysis of 11 varied studies, with a search in various databases and a rigorous selection of relevant studies. The results reveal the promising performance of GPT models, but also highlight limitations such as their inability to understand nuances and complex questions. In addition, challenges persist in terms of intellectual property, confidentiality and reliability of responses. The studies include performance evaluations, narrative reviews, a systematic review and a case study on the implementation of a ChatGPT-based chatbot in a university library. These studies offer interesting insights into the potential benefits and challenges of successfully integrating ChatGPT into library services. [ABSTRACT FROM AUTHOR]

In this study, 100 self-reported personality items from the Big Five Aspects Scale, responded to by a sample of 334 undergraduate participants, were used to predict quantity (ideational fluency) and quality (originality) of ideas on a divergent thinking (DT) task. The originality of DT responses was scored through a fine-tuned version of the Generative Pre-trained Transformer (GPT) 3.5 (i.e., Ocsai), and a least absolute shrinkage selection operator (LASSO) machine learning model selected the items that were meaningful predictors of each outcome. Results revealed that the personality profiles of highly fluent and highly original individuals were characterized by a tension between seemingly opposed personality attributes. Both ideational fluency and originality were predicted by a playfully open intellectualism that nonetheless avoided more typical work (i.e. was disorderly and unindustrious). Fluency was additionally predicted by a tension between enthusiasm for social interaction and depressive symptoms associated with withdrawal. Originality was predicted by a socially dominant assertiveness that was tempered by awareness and care for others’ feelings (e.g. compassion and politeness) as well as stability (i.e. non-volatility). Taken together, these results demonstrate that the creative personality is likely to be composed of aspects of multiple dimensions of typical personality models like the Big 5, and that the highly fluent and the highly original creative personality is different in important ways. [ABSTRACT FROM AUTHOR]

[Objectives] Aiming at the lack of disease description in rice diseases and pests image classification technology, a lightweight rice diseases and pests image description model was proposed in this paper to describe rice diseases and pests image more specifically. [Methods] Ten common rice pests and diseases, such as rice bacterial blight, rice bacterial streak disease, rice bakanae disease, rice three chemical borers, rice blast, rice false smut, rice sheath blight, rice planthopper, rice thrip and rice brown spot were studied, and Chinese description data set of rice pests and diseases image was constructed. Firstly, the multimodal pre-training model CLIP was used to generate image vectors, which contained basic image information and rich semantic information. The mapping network was used to map the image vectors into the text space to generate text prompt vectors. Finally, the language model GPT-2 generates image descriptions according to the prompt vectors. [Results] The test results showed that the indexes of the model in this paper were significantly superior to other models in the image description data set of rice pests and diseases. Compared with the traditional CNN_ LSTM model, the indexes of BLEU-1, BLEU-2, BLEU-3, BLEU-4, ROUGE and METEOR improved 0.26, 0.27, 0.24, 0.22, 0.22 and 0.14, respectively. And the generated image description had the advantages of accurate, detailed and rich semantics. The model was tested by using actual rice field pictures. The actual field scenes were more complex and diverse, and the generated image description index only slightly decreased compared with the data set index, which was still higher than other comparison models. The overall recognition accuracy of the model was 97.28%. [Conclusions] The image description method of rice diseases and pests based on multimodal pre-training model can accurately describe the rice diseases and pests, and provide a new idea for the detection of rice diseases and pests. [ABSTRACT FROM AUTHOR]

Systematic reviews are vital for guiding practice, research and policy, although they are often slow and labour‐intensive. Large language models (LLMs) could speed up and automate systematic reviews, but their performance in such tasks has yet to be comprehensively evaluated against humans, and no study has tested Generative Pre‐Trained Transformer (GPT)‐4, the biggest LLM so far. This pre‐registered study uses a "human‐out‐of‐the‐loop" approach to evaluate GPT‐4's capability in title/abstract screening, full‐text review and data extraction across various literature types and languages. Although GPT‐4 had accuracy on par with human performance in some tasks, results were skewed by chance agreement and dataset imbalance. Adjusting for these caused performance scores to drop across all stages: for data extraction, performance was moderate, and for screening, it ranged from none in highly balanced literature datasets (~1:1) to moderate in those datasets where the ratio of inclusion to exclusion in studies was imbalanced (~1:3). When screening full‐text literature using highly reliable prompts, GPT‐4's performance was more robust, reaching "human‐like" levels. Although our findings indicate that, currently, substantial caution should be exercised if LLMs are being used to conduct systematic reviews, they also offer preliminary evidence that, for certain review tasks delivered under specific conditions, LLMs can rival human performance. [ABSTRACT FROM AUTHOR]

Background and Aims: The Liver Imaging Reporting and Data System (LI-RADS) offers a standardized approach for imaging hepatocellular carcinoma. However, the diverse styles and structures of radiology reports complicate automatic data extraction. Large language models hold the potential for structured data extraction from free-text reports. Our objective was to evaluate the performance of Generative Pre-trained Transformer (GPT)-4 in extracting LI-RADS features and categories from free-text liver magnetic resonance imaging (MRI) reports. Methods: Three radiologists generated 160 fictitious free-text liver MRI reports written in Korean and English, simulating real-world practice. Of these, 20 were used for prompt engineering, and 140 formed the internal test cohort. Seventy-two genuine reports, authored by 17 radiologists were collected and de-identified for the external test cohort. LI-RADS features were extracted using GPT-4, with a Python script calculating categories. Accuracies in each test cohort were compared. Results: On the external test, the accuracy for the extraction of major LI-RADS features, which encompass size, nonrim arterial phase hyperenhancement, nonperipheral 'washout', enhancing 'capsule' and threshold growth, ranged from .92 to .99. For the rest of the LI-RADS features, the accuracy ranged from .86 to .97. For the LI-RADS category, the model showed an accuracy of .85 (95% CI: .76, .93). Conclusions: GPT-4 shows promise in extracting LI-RADS features, yet further refinement of its prompting strategy and advancements in its neural network architecture are crucial for reliable use in processing complex real-world MRI reports. [ABSTRACT FROM AUTHOR]

AbstractResearch consistently shows the importance of teacher and peer feedback in enabling students to generate internal feedback for improved learning. However, the utilisation of Generative Pre-Trained Transformer (ChatGPT) to support higher education students in generating internal feedback lacks clarity. This qualitative case study aimed to explore students’ experiences with ChatGPT as they sought assistance in generating internal feedback and to identify factors that influence the quality of feedback provided by ChatGPT. Forty-eight Chinese students with varying levels of writing proficiency participated in an English writing course, and data were collected through self-reflective questions, interviews, and conversation logs with ChatGPT. Qualitative data were analyzed using thematic coding. This qualitative case study aimed to explore students’ experiences with ChatGPT as they sought assistance in generating internal feedback and to identify the factors that influence the quality of feedback provided by ChatGPT. The quality of ChatGPT’s feedback was influenced by factors such as learners’ personal goals, the utilisation of the system’s multilingual capability, and their self-regulation writing strategies, which in turn affect the quality of prompts. These findings offer practical insights into the use of ChatGPT for generating internal feedback in second language (L2) writing. [ABSTRACT FROM AUTHOR]

Released in March 2023, GPT‐4 has shown remarkable proficiency in various scientific areas, including atmospheric pressure plasma research. The paper discusses how GPT‐4, supported by ScholarAI and Wolfram plugins, enhances research efficiency with its vast knowledge and inferencing skills. It focuses on plasma‐water‐based nitrogen fixation experiments, demonstrating GPT‐4's abilities in explanation, analysis, prediction, planning, and verification. The study also addresses challenges like interpreting non‐textual data and keeping pace with scientific advancements. It highlights GPT‐4's role in improving nitrogen fixation process efficiency, showcasing its utility in scientific analysis and decision‐making. The paper emphasizes the evolving role of AI in research and the need for the scientific community to adapt to the rise of more advanced AI systems. [ABSTRACT FROM AUTHOR]

Large language models are capable of generating source code in a zero-shot manner to develop programs that meet user functional requirements. However, when faced with scenarios involving complex business requirements, the generated source code may fail to satisfy user needs. Addressing the challenge of understanding software requirements, we propose a two-stage code generation approach. Initially, the large language model generates pseudocode based on the user's functional requirements, refined through an iterative process with user feedback. Subsequently, the model generates source code based on the finalized pseudocode. We conducted empirical studies on an open code generation dataset, and experimental results with models such as GPT-4, Claude Sonnect 3, and Geminipro 1.5 demonstrate that our method outperforms zero-shot prompt learning in scenarios with complex user requirements, with improvements in PASS@K reaching up to 15%. [ABSTRACT FROM AUTHOR]

The springing up of large language models (LLMs) has shifted the community from single-task-orientated natural language processing (NLP) research to a holistic end-to-end multi-task learning paradigm. Along this line of research endeavors in the area, LLM-based prompting methods have attracted much attention, partially due to the technological advantages brought by prompt engineering (PE) as well as the underlying NLP principles disclosed by various prompting methods. Traditional supervised learning usually requires training a model based on labeled data and then making predictions. In contrast, PE methods directly use the powerful capabilities of existing LLMs (e.g., GPT-3 and GPT-4) via composing appropriate prompts, especially under few-shot or zero-shot scenarios. Facing the abundance of studies related to the prompting and the ever-evolving nature of this field, this article aims to 1) illustrate a novel perspective to review existing PE methods within the well-established communication theory framework, 2) facilitate a better/deeper understanding of developing trends of existing PE methods used in three typical tasks, and 3) shed light on promising research directions for future PE methods. [ABSTRACT FROM AUTHOR]

This paper examines the potential of large language models to mimic human survey respondents and to derive their preferences. We explore the viability of large language models (LLMs), specifically OpenAI's GPT-3.5 and GPT-4, in emulating human survey respondents and eliciting preferences, with a focus on intertemporal choices. Leveraging the extensive literature on intertemporal discounting for benchmarking, we examine responses from LLMs across various languages and compare them with human responses, exploring preferences between smaller, sooner and larger, later rewards. Our findings reveal that both generative pretrained transformer (GPT) models demonstrate less patience than humans, with GPT-3.5 exhibiting a lexicographic preference for earlier rewards unlike human decision makers. Although GPT-4 does not display lexicographic preferences, its measured discount rates are still considerably larger than those found in humans. Interestingly, GPT models show greater patience in languages with weak future tense references, such as German and Mandarin, aligning with the existing literature that suggests a correlation between language structure and intertemporal preferences. We demonstrate how prompting GPT to explain its decisions, a procedure we term "chain-of-thought conjoint," can mitigate, but does not eliminate, discrepancies between LLM and human responses. Although directly eliciting preferences using LLMs may yield misleading results, combining chain-of-thought conjoint with topic modeling aids in hypothesis generation, enabling researchers to explore the underpinnings of preferences. Chain-of-thought conjoint provides a structured framework for marketers to use LLMs to identify potential attributes or factors that can explain preference heterogeneity across different customers and contexts. History: Olivier Toubia served as the senior editor. This paper was accepted through the Marketing Science: Frontiers review process. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mksc.2023.0306. [ABSTRACT FROM AUTHOR]

Copyright of International Journal of Educational Psychological Studies (EPS) is the property of Refaad for Studies, Research & Development 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.)

This work explores the mathematical modelling capabilities of various iterations of ChatGPT, focusing on their performance across tasks of differing complexity and openness. The study examines the abilities of GPT-3.5, GPT-4.0, and a more instructed version, GPT-MM, in multiple scenarios. It is observed that all versions demonstrate basic mathematical problem-solving skills. However, their effectiveness varies with increasing task complexity. While GPT-4.0 and GPT-MM show marginal improvements in providing detailed solutions, significant challenges persist, especially in moderate to complex modelling contexts where comprehending the nuances of tasks becomes challenging. Additionally, the study suggests that the openness of modelling tasks has a limited impact on performance, highlighting that mathematical and contextual complexities play more critical roles. The implications of these observations are discussed in terms of potential enhancements to teaching methodologies and the integration of AI tools like GPT in educational settings. This reiterates the importance of further research to fully understand the capabilities and limitations of AI tools and ensure their effective use in education. [ABSTRACT FROM AUTHOR]

This study analyzes pretrained language models’ ability to distinguish between literal and figurative meanings of Korean somatosensory figurative expressions. We created an experimental database with 27 expressions, each paired with sentences demonstrating literal and figurative uses. Using GPT-3.5, GPT-4, and GPT-4-Opus, we conducted zero-shot learning tasks to determine the usage in each sentence. GPT-4 and GPT-4-Opus achieved 100% accuracy, while GPT-3.5 reached 92.59%. Additional one-shot and two-shot learning experiments with GPT-3.5 resulted in decreased performance at 74.07% and 81.48% respectively, highlighting the complexity of few-shot learning for this task. This study confirms that state-of-the-art language models can understand somatosensory figurative expressions with high accuracy, emphasizes the importance of prompt design, and reveals limitations of few-shot learning in this context. These findings suggest possibilities for computational validation of embodied cognition theory and contribute to advancing language models and natural language processing techniques for figurative expression processing. [ABSTRACT FROM AUTHOR]

This study investigates the efficacy of advanced large language models, specifically GPT-4o, Claude-3.5 Sonnet, and GPT-3.5 Turbo, in detecting software vulnerabilities. Our experiment utilized vulnerable and secure code samples from the NIST Software Assurance Reference Dataset (SARD), focusing on C++, Java, and Python. We employed three distinct prompting techniques as follows: Concise, Tip Setting, and Step-by-Step. The results demonstrate that GPT-4o and Claude-3.5 Sonnet significantly outperform GPT-3.5 Turbo in vulnerability detection. GPT-4o showed the highest improvement with the Step-by-Step prompt, achieving an F1 score of 0.9072. Claude-3.5 Sonnet exhibited consistent high performance across all prompt types, with its Step-by-Step prompt yielding the best overall results (F1 score: 0.8933, AUC: 0.74). In contrast, GPT-3.5 Turbo showed minimal performance changes across prompts, with the Tip Setting prompt performing best (AUC: 0.65, F1 score: 0.6772), yet significantly lower than the other models. Our findings highlight the potential of advanced models in enhancing software security and underscore the importance of prompt engineering in optimizing their performance. [ABSTRACT FROM AUTHOR]

The body of ecological literature, which informs much of our knowledge of the global loss of biodiversity, has been experiencing rapid growth in recent decades. The increasing difficulty of synthesising this literature manually has simultaneously resulted in a growing demand for automated text mining methods. Within the domain of deep learning, large language models (LLMs) have been the subject of considerable attention in recent years due to great leaps in progress and a wide range of potential applications; however, quantitative investigation into their potential in ecology has so far been lacking.In this work, we analyse the ability of GPT‐4 to extract information about invertebrate pests and pest controllers from abstracts of articles on biological pest control, using a bespoke, zero‐shot prompt.Our results show that the performance of GPT‐4 is highly competitive with other state‐of‐the‐art tools used for taxonomic named entity recognition and geographic location extraction tasks. On a held‐out test set, we show that species and geographic locations are extracted with F1‐scores of 99.8% and 95.3%, respectively, and highlight that the model can effectively distinguish between ecological roles of interest such as predators, parasitoids and pests. Moreover, we demonstrate the model's ability to effectively extract and predict taxonomic information across various taxonomic ranks. However, we do report a small number of cases of fabricated information (confabulations).Due to a lack of specialised, pre‐trained ecological language models, general‐purpose LLMs may provide a promising way forward in ecology. Combined with tailored prompt engineering, such models can be employed for a wide range of text mining tasks in ecology, with the potential to greatly reduce time spent on manual screening and labelling of the literature. [ABSTRACT FROM AUTHOR]

The recent big model such as GPT-3.5 possesses an extensive understanding of natural language, and it can perform a wide range of tasks, making it a significant advancement in the field of artificial intelligence (AI). A critical challenge in the design and implementation of big model is that it imposes a heavy load on the wireless transmission due to a huge size of the deep network parameters, especially for the distributed implementation. To tackle this challenge, we investigate big model transmission under practical double Rayleigh fading environments, where the big model is simultaneously distributed to multiple training nodes through wireless transmission. To evaluate the system performance, we study the system outage probability (OP) based on the transmission latency, where an analytical expression is derived for the OP, which is used to evaluate the performance of big model transmission under double Rayleigh fading environments. Finally, we present some simulations under double Rayleigh fading environments, in order to show the validity of the proposed big model transmission. [ABSTRACT FROM AUTHOR]

Despite the widespread adoption of generative transformer large language models and the interest of the global legal community, discussions about the models in philosophy of law mainly have been focusing on what LLMs cannot do. In making the first steps towards a philosophical analysis of the capabilities of AI models in the field of law, we follow the basic idea of Turing's „imitation game“. Proceeding from the frequently raised characterization of legal reasoning as „artificial“, the paper identifies the undisputed minimum core of the „artificiality“ thesis and asks to what extent it can be imitated by artificial intelligence. To answer this question, we test the legal reasoning capabilities of ChatGPT, the most advanced, up-to-date LLM version of artificial intelligence. The conclusion is that in all relevant types of activities usually associated with legal reasoning - fact-finding, interpretation, qualification, and decision-making - ChatGPT can generate outcomes as if it reasons legally. [ABSTRACT FROM AUTHOR]

Recent advancements in natural language processing (NLP) and large language models (LLMs) have the potential to greatly enhance model-informed drug development (MIDD). LLMs, with their billions of parameters and pre-training on diverse data, can effectively extract information from unstructured and structured data throughout the drug development process. These models offer unique capabilities such as prompt engineering, which allows them to complete tasks based on prompts without further training. Implementing LLMs in MIDD can improve data curation, data cleaning, knowledge extraction, and data discovery processes. However, there are challenges to address, including the usability of LLMs, the need for effective prompt design, and concerns about interpretability and bias. Continued open-science development and evaluation of LLMs will be crucial to fully utilize their potential in MIDD. [Extracted from the article]

Purpose: Large Language Models (LLMs) like Generative Pre-trained Transformer (GPT) from OpenAI and LLaMA (Large Language Model Meta AI) from Meta AI are increasingly recognized for their potential in the field of cheminformatics, particularly in understanding Simplified Molecular Input Line Entry System (SMILES), a standard method for representing chemical structures. These LLMs also have the ability to decode SMILES strings into vector representations. Method: We investigate the performance of GPT and LLaMA compared to pre-trained models on SMILES in embedding SMILES strings on downstream tasks, focusing on two key applications: molecular property prediction and drug-drug interaction prediction. Results: We find that SMILES embeddings generated using LLaMA outperform those from GPT in both molecular property and DDI prediction tasks. Notably, LLaMA-based SMILES embeddings show results comparable to pre-trained models on SMILES in molecular prediction tasks and outperform the pre-trained models for the DDI prediction tasks. Conclusion: The performance of LLMs in generating SMILES embeddings shows great potential for further investigation of these models for molecular embedding. We hope our study bridges the gap between LLMs and molecular embedding, motivating additional research into the potential of LLMs in the molecular representation field. GitHub: https://github.com/sshaghayeghs/LLaMA-VS-GPT. [ABSTRACT FROM AUTHOR]

De novo drug design is crucial in advancing drug discovery, which aims to generate new drugs with specific pharmacological properties. Recently, deep generative models have achieved inspiring progress in generating drug-like compounds. However, the models prioritize a single target drug generation for pharmacological intervention, neglecting the complicated inherent mechanisms of diseases, and influenced by multiple factors. Consequently, developing novel multi-target drugs that simultaneously target specific targets can enhance anti-tumor efficacy and address issues related to resistance mechanisms. To address this issue and inspired by Generative Pre-trained Transformers (GPT) models, we propose an upgraded GPT model with generative adversarial imitation learning for multi-target molecular generation called MTMol-GPT. The multi-target molecular generator employs a dual discriminator model using the Inverse Reinforcement Learning (IRL) method for a concurrently multi-target molecular generation. Extensive results show that MTMol-GPT generates various valid, novel, and effective multi-target molecules for various complex diseases, demonstrating robustness and generalization capability. In addition, molecular docking and pharmacophore mapping experiments demonstrate the drug-likeness properties and effectiveness of generated molecules potentially improve neuropsychiatric interventions. Furthermore, our model's generalizability is exemplified by a case study focusing on the multi-targeted drug design for breast cancer. As a broadly applicable solution for multiple targets, MTMol-GPT provides new insight into future directions to enhance potential complex disease therapeutics by generating high-quality multi-target molecules in drug discovery. Author summary: Advancing drug discovery with the vast scale of possible structures of drug-like compounds (between 1023 and 1060) relies on Artificial Intelligence (AI). The process of generating SMILES has been accomplished by using language models like the Generative Pre-trained Transformer (GPT) model. In the meantime, regulating multiple targets to achieve the desired physiological responses is booming for treating complicated diseases. However, the recent deep generative models focus on single-target drug generation, and single-target drugs have limitations, such as requiring higher doses and causing mutations. For example, activation of Src kinase in certain breast cancer cells leads to EGFR phosphorylation and downstream effects. Consequently, designing multi-target molecules against multi-target simultaneously is highly required. Thus, we have proposed an upgraded GPT model MTMol-GPT with generative adversarial imitation learning for generating high-quality multiple targets drug-like compounds. For example, MTMol-GPT generates novel drugs targeting protein pairs such as DRD2 and HTR1A, EGFR and Src. As a broadly applicable tool, MTMol-GPT provides a rapid and accurate method for advancing drug discovery, ensuring the efficient generation of high-quality multi-target drugs. [ABSTRACT FROM AUTHOR]

Background: Large language models (LLMs) that can efficiently screen and identify studies meeting specific criteria would streamline literature reviews. Additionally, those capable of extracting data from publications would enhance knowledge discovery by reducing the burden on human reviewers. Methods: We created an automated pipeline utilizing OpenAI GPT-4 32 K API version "2023–05-15" to evaluate the accuracy of the LLM GPT-4 responses to queries about published papers on HIV drug resistance (HIVDR) with and without an instruction sheet. The instruction sheet contained specialized knowledge designed to assist a person trying to answer questions about an HIVDR paper. We designed 60 questions pertaining to HIVDR and created markdown versions of 60 published HIVDR papers in PubMed. We presented the 60 papers to GPT-4 in four configurations: (1) all 60 questions simultaneously; (2) all 60 questions simultaneously with the instruction sheet; (3) each of the 60 questions individually; and (4) each of the 60 questions individually with the instruction sheet. Results: GPT-4 achieved a mean accuracy of 86.9% – 24.0% higher than when the answers to papers were permuted. The overall recall and precision were 72.5% and 87.4%, respectively. The standard deviation of three replicates for the 60 questions ranged from 0 to 5.3% with a median of 1.2%. The instruction sheet did not significantly increase GPT-4's accuracy, recall, or precision. GPT-4 was more likely to provide false positive answers when the 60 questions were submitted individually compared to when they were submitted together. Conclusions: GPT-4 reproducibly answered 3600 questions about 60 papers on HIVDR with moderately high accuracy, recall, and precision. The instruction sheet's failure to improve these metrics suggests that more sophisticated approaches are necessary. Either enhanced prompt engineering or finetuning an open-source model could further improve an LLM's ability to answer questions about highly specialized HIVDR papers. [ABSTRACT FROM AUTHOR]

Purpose: To examine the ability of generative artificial intelligence (GAI) to answer patients' questions regarding colorectal cancer (CRC). Methods: Ten clinically relevant questions about CRC were selected from top-rated hospitals' websites and patient surveys and presented to three GAI tools (Chatbot Generative Pre-Trained Transformer [GPT-4], Google Bard, and CLOVA X). Their responses were compared with answers from the CRC information book. Response evaluation was performed by two groups, each consisting of five healthcare professionals (HCP) and patients. Each question was scored on a 1–5 Likert scale based on four evaluation criteria (maximum score, 20 points/question). Results: In an analysis including only HCPs, the information book scored 11.8 ± 1.2, GPT-4 scored 13.5 ± 1.1, Google Bard scored 11.5 ± 0.7, and CLOVA X scored 12.2 ± 1.4 (P = 0.001). The score of GPT-4 was significantly higher than those of the information book (P = 0.020) and Google Bard (P = 0.001). In an analysis including only patients, the information book scored 14.1 ± 1.4, GPT-4 scored 15.2 ± 1.8, Google Bard scored 15.5 ± 1.8, and CLOVA X scored 14.4 ± 1.8, without significant differences (P = 0.234). When both groups of evaluators were included, the information book scored 13.0 ± 0.9, GPT-4 scored 14.4 ± 1.2, Google Bard scored 13.5 ± 1.0, and CLOVA X scored 13.3 ± 1.5 (P = 0.070). Conclusion: The three GAIs demonstrated similar or better communicative competence than the information book regarding questions related to CRC surgery in Korean. If high-quality medical information provided by GAI is supervised properly by HCPs and published as an information book, it could be helpful for patients to obtain accurate information and make informed decisions. [ABSTRACT FROM AUTHOR]

AbstractTo address the long-standing challenge facing traditional automated writing evaluation (AWE) systems in assessing higher-order thinking, this study built an AWE system for scoring argumentative essays by finetuning the GPT-3.5 Large Language Model. The system’s effectiveness was compared with that of the non-finetuned GPT-3.5 and GPT-4 base models via zero-shot prompting, which involves applying the model to perform tasks without any prior specific training or examples on those tasks. The dataset used was the TOEFL Public Writing Dataset provided by Education Testing Service (ETS), containing 480 argumentative essays with ground truth scores under two essay prompts. Three finetuned models were generated: two finetuned exclusively on either prompt and one on both. All finetuned and base models were used to score the remaining essays after finetuning and their scoring effectiveness was compared with ground truth scores, i.e., benchmark scores assigned by ETS-trained human raters. The impact of the variety of finetuning prompts and the robustness of finetuned models were also explored. Results showed a 100% consistency of all models in two scoring sessions. More importantly, the finetuned models significantly outperformed the base models in accuracy and reliability. The best-performing model, finetuned on prompt 1, showed an RMSE of 0.57, a percentage agreement (score discrepancy ≤ 0.5) of 84.72% and a QWK of 0.78. Further, the model finetuned on both prompts did not exhibit enhanced performance, and the two models finetuned on one prompt remained robust when scoring essays from the alternative prompt. These results suggest (1) task-specific finetuning for AWE is beneficial; (2) finetuning does not require a large variety of essay prompts; and (3) fine-tuned models are robust to unseen essays. [ABSTRACT FROM AUTHOR]

GPT-4, as the most advanced version of OpenAI's large language models, has attracted widespread attention, rapidly becoming an indispensable AI tool across various areas. This includes its exploration by scientists for diverse applications. Our study focused on assessing GPT-4's capabilities in generating text, tables, and diagrams for biomedical review papers. We also assessed the consistency in text generation by GPT-4, along with potential plagiarism issues when employing this model for the composition of scientific review papers. Based on the results, we suggest the development of enhanced functionalities in ChatGPT, aiming to meet the needs of the scientific community more effectively. This includes enhancements in uploaded document processing for reference materials, a deeper grasp of intricate biomedical concepts, more precise and efficient information distillation for table generation, and a further refined model specifically tailored for scientific diagram creation. [ABSTRACT FROM AUTHOR]

Symbolic task planning is a widely used approach to enforce robot autonomy due to its ease of understanding and deployment in engineered robot architectures. However, techniques for symbolic task planning are difficult to scale in real-world, highly dynamic, human-robot collaboration scenarios because of the poor performance in planning domains where action effects may not be immediate, or when frequent re-planning is needed due to changed circumstances in the robot workspace. The validity of plans in the long term, plan length, and planning time could hinder the robot's efficiency and negatively affect the overall human-robot interaction's fluency. We present a framework, which we refer to as Teriyaki, specifically aimed at bridging the gap between symbolic task planning and machine learning approaches. The rationale is training Large Language Models (LLMs), namely GPT-3, into a neurosymbolic task planner compatible with the Planning Domain Definition Language (PDDL), and then leveraging its generative capabilities to overcome a number of limitations inherent to symbolic task planners. Potential benefits include (i) a better scalability in so far as the planning domain complexity increases, since LLMs' response time linearly scales with the combined length of the input and the output, instead of super-linearly as in the case of symbolic task planners, and (ii) the ability to synthesize a plan action-by-action instead of end-to-end, and to make each action available for execution as soon as it is generated instead of waiting for the whole plan to be available, which in turn enables concurrent planning and execution. In the past year, significant efforts have been devoted by the research community to evaluate the overall cognitive capabilities of LLMs, with alternate successes. Instead, with Teriyaki we aim to providing an overall planning performance comparable to traditional planners in specific planning domains, while leveraging LLMs capabilities in other metrics, specifically those related to their short- and mid-term generative capabilities, which are used to build a look-ahead predictive planning model. Preliminary results in selected domains show that our method can: (i) solve 95.5% of problems in a test data set of 1,000 samples; (ii) produce plans up to 13.5% shorter than a traditional symbolic planner; (iii) reduce average overall waiting times for a plan availability by up to 61.4%. [ABSTRACT FROM AUTHOR]

Small-molecule drug design aims to generate compounds that target specific proteins, playing a crucial role in the early stages of drug discovery. Recently, research has emerged that utilizes the GPT model, which has achieved significant success in various fields to generate molecular compounds. However, due to the persistent challenge of small datasets in the pharmaceutical field, there has been some degradation in the performance of generating target-specific compounds. To address this issue, we propose an enhanced target-specific drug generation model, Adapt-cMolGPT, which modifies molecular representation and optimizes the fine-tuning process. In particular, we introduce a new fine-tuning method that incorporates an adapter module into a pre-trained base model and alternates weight updates by sections. We evaluated the proposed model through multiple experiments and demonstrated performance improvements compared to previous models. In the experimental results, Adapt-cMolGPT generated a greater number of novel and valid compounds compared to other models, with these generated compounds exhibiting properties similar to those of real molecular data. These results indicate that our proposed method is highly effective in designing drugs targeting specific proteins. [ABSTRACT FROM AUTHOR]

The rapid evolution of large language models, in particular OpenAI's GPT-3.5-turbo and GPT-4, indicates a growing interest in advanced computational methodologies. This paper proposes a novel approach to synthetic data generation and knowledge distillation through prompt engineering. The potential of large language models (LLMs) is used to address the problem of unbalanced training datasets for other machine learning models. This is not only a common issue but also a crucial determinant of the final model quality and performance. Three prompting strategies have been considered: basic, composite, and similarity prompts. Although the initial results do not match the performance of comprehensive datasets, the similarity prompts method exhibits considerable promise, thus outperforming other methods. The investigation of our rebalancing methods opens pathways for future research on leveraging continuously developed LLMs for the enhanced generation of high-quality synthetic data. This could have an impact on many large-scale engineering applications. [ABSTRACT FROM AUTHOR]

Recent advancements in large language models (LLMs) have raised the prospect of scalable, automated, and fine-grained political microtargeting on a scale previously unseen; however, the persuasive influence of microtargeting with LLMs remains unclear. Here, we build a custom web application capable of integrating self-reported demographic and political data into GPT-4 prompts in real-time, facilitating the live creation of unique messages tailored to persuade individual users on four political issues. We then deploy this application in a preregistered randomized control experiment (n = 8,587) to investigate the extent to which access to individual-level data increases the persuasive influence of GPT-4. Our approach yields two key findings. First, messages generated by GPT-4 were broadly persuasive, in some cases increasing support for an issue stance by up to 12 percentage points. Second, in aggregate, the persuasive impact of microtargeted messages was not statistically different from that of non-microtargeted messages (4.83 vs. 6.20 percentage points, respectively, P = 0.226). These trends hold even when manipulating the type and number of attributes used to tailor the message. These findings suggest--contrary to widespread speculation--that the influence of current LLMs may reside not in their ability to tailor messages to individuals but rather in the persuasiveness of their generic, nontargeted messages. We release our experimental dataset, GPTarget2024, as an empirical baseline for future research. [ABSTRACT FROM AUTHOR]

There is much excitement about the opportunity to harness the power of large language models (LLMs) when building problem-solving assistants. However, the standard methodology of evaluating LLMs relies on static pairs of inputs and outputs; this is insufficient for making an informed decision about which LLMs are best to use in an interactive setting, and how that varies by setting. Static assessment therefore limits how we understand language model capabilities. We introduce CheckMate, an adaptable prototype platform for humans to interact with and evaluate LLMs. We conduct a study with CheckMate to evaluate three language models (InstructGPT, ChatGPT, and GPT-4) as assistants in proving undergraduate-level mathematics, with a mixed cohort of participants from undergraduate students to professors of mathematics. We release the resulting interaction and rating dataset, MathConverse. By analyzing MathConverse, we derive a taxonomy of human query behaviors and uncover that despite a generally positive correlation, there are notable instances of divergence between correctness and perceived helpfulness inLLMgenerations, among other findings. Further, we garner a more granular understanding of GPT-4 mathematical problemsolving through a series of case studies, contributed by experienced mathematicians. We conclude with actionable takeaways for ML practitioners and mathematicians: models that communicate uncertainty, respond well to user corrections, and can provide a concise rationale for their recommendations, may constitute better assistants. Humans should inspect LLM output carefully given their current shortcomings and potential for surprising fallibility. [ABSTRACT FROM AUTHOR]

Large language models (LLMs) are currently at the forefront of intertwining AI systems with human communication and everyday life. Thus, aligning them with human values is of great importance. However, given the steady increase in reasoning abilities, future LLMs are under suspicion of becoming able to deceive human operators and utilizing this ability to bypass monitoring efforts. As a prerequisite to this, LLMs need to possess a conceptual understanding of deception strategies. This study reveals that such strategies emerged in state-of-the-art LLMs, but were nonexistent in earlier LLMs. We conduct a series of experiments showing that state-of-the-art LLMs are able to understand and induce false beliefs in other agents, that their performance in complex deception scenarios can be amplified utilizing chain-of-thought reasoning, and that eliciting Machiavellianism in LLMs can trigger misaligned deceptive behavior. GPT-4, for instance, exhibits deceptive behavior in simple test scenarios 99.16% of the time (P < 0.001). In complex second-order deception test scenarios where the aim is to mislead someone who expects to be deceived, GPT-4 resorts to deceptive behavior 71.46% of the time (P < 0.001) when augmented with chain-of-thought reasoning. In sum, revealing hitherto unknown machine behavior in LLMs, our study contributes to the nascent field of machine psychology. [ABSTRACT FROM AUTHOR]

This comprehensive exploration of recent breakthroughs in artificial intelligence (AI) traversed the realms of language models, computer vision, and generative models, unraveling the intricacies of cutting-edge technologies such as GPT-3.5, GPT-4, Pix2Seq, and multimodal models in terms of generative AI. In this multifaceted journey, the focus extended beyond technological prowess to ethical considerations, emphasizing responsible AI practices guided by Google's AI Principles. The nuanced discussions encapsulated the transformative impact of AI on user experiences across various Google products and toolsets, paving the way for a future where natural language interaction, creative content generation, and multimodal understanding redefine human–computer interactions. The research investigation showcased not only the advancements themselves but also the critical lens through which these innovations are approached, underscoring the importance of ethical and responsible AI in shaping the technological landscape. [ABSTRACT FROM AUTHOR]

The computational analysis of human personality has mainly focused on the Big Five personality theory, and the psychodynamic approach is almost nonexistent despite its rich theoretical grounding and relevance to various tasks. Here, we provide a data set of 4972 synthetic utterances corresponding with five personality dimensions described by the psychodynamic approach: depressive, obsessive, paranoid, narcissistic, and anti-social psychopathic. The utterances have been generated through AI with a deep theoretical orientation that motivated the design of prompts for GPT-4. The dataset has been validated through 14 tests, and it may be relevant for the computational study of human personality and the design of authentic persona in digital domains, from gaming to the artistic generation of movie characters. [ABSTRACT FROM AUTHOR]

Motivation Macrocyclic peptides hold great promise as therapeutics targeting intracellular proteins. This stems from their remarkable ability to bind flat protein surfaces with high affinity and specificity while potentially traversing the cell membrane. Research has already explored their use in developing inhibitors for intracellular proteins, such as KRAS, a well-known driver in various cancers. However, computational approaches for de novo macrocyclic peptide design remain largely unexplored. Results Here, we introduce HELM-GPT, a novel method that combines the strength of the hierarchical editing language for macromolecules (HELM) representation and generative pre-trained transformer (GPT) for de novo macrocyclic peptide design. Through reinforcement learning (RL), our experiments demonstrate that HELM-GPT has the ability to generate valid macrocyclic peptides and optimize their properties. Furthermore, we introduce a contrastive preference loss during the RL process, further enhanced the optimization performance. Finally, to co-optimize peptide permeability and KRAS binding affinity, we propose a step-by-step optimization strategy, demonstrating its effectiveness in generating molecules fulfilling both criteria. In conclusion, the HELM-GPT method can be used to identify novel macrocyclic peptides to target intracellular proteins. Availability and implementation The code and data of HELM-GPT are freely available on GitHub (https://github.com/charlesxu90/helm-gpt). [ABSTRACT FROM AUTHOR]

The use of artificial intelligence (AI) tools to aid in summarizing information in medicine and research has recently garnered a huge amount of interest. While tools such as ChatGPT produce convincing and naturally sounding output, the answers are sometimes incorrect. Some of these drawbacks, it is hoped, can be avoided by using programmes trained for a more specific scope. In this study we compared the performance of a new AI tool (the-literature.com) to the latest version OpenAI's ChatGPT (GPT-4) in summarizing topics that the authors have significantly contributed to. The AI tools were asked to produce a literature review on 7 topics. These were selected based on the research topics that the authors were intimately familiar with and have contributed to through their own publications. The output produced by the AI tools were graded on a 1–5 Likert scale for accuracy, comprehensiveness, and relevance by two fellowship trained consultant radiologists. The-literature.com produced 3 excellent summaries, 3 very poor summaries not relevant to the prompt, and one summary, which was relevant but did not include all relevant papers. All of the summaries produced by GPT-4 were relevant, but fewer relevant papers were identified. The average Likert rating was for the-literature was 2.88 and 3.86 for GPT-4. There was good agreement between the ratings of both radiologists (ICC = 0.883). Summaries produced by AI in its current state require careful human validation. GPT-4 on average provides higher quality summaries. Neither tool can reliably identify all relevant publications. • Literature review is a crucial, but labour intensive, part of research. • Artificial intelligence tools have the potential to realise significant time savings. • AI driven tools can in certain cases produce impressive literature reviews. • However, current AI literature review tools are too unreliable for routine use. [ABSTRACT FROM AUTHOR]

This paper introduces the collection of the Journal on Machine Learning (ML) and the user. It provides a brief history of ML from the 1950's through to the current time, sketching the nature of the kinds of precursor AI techniques used in such things as expert systems right the way through to the emergence of ML and its tool sets, including deep learning. It concludes with the 'generative AI' used in such ML technologies as PaLM and GPT-3. The history highlights key changes and developments in ML, the especial importance and limitations of deep learning, and the changing attitudes and expectations of users in an environment when ML can and often is oversold. The paper then explores the ways CSCW research has addressed the social context of organisational systems and how the same can apply for ML tools and techniques. It urges research that focuses on the particular ways that ML comes to fit into 'real world' collaborative work sites and hence speaks to the CSCW cannon. [ABSTRACT FROM AUTHOR]

PURPOSE: Artificial intelligence (AI) models can generate scientific abstracts that are difficult to distinguish from the work of human authors. The use of AI in scientific writing and performance of AI detection tools are poorly characterized. METHODS: We extracted text from published scientific abstracts from the ASCO 2021-2023 Annual Meetings. Likelihood of AI content was evaluated by three detectors: GPTZero, Originality.ai, and Sapling. Optimal thresholds for AI content detection were selected using 100 abstracts from before 2020 as negative controls, and 100 produced by OpenAI's GPT-3 and GPT-4 models as positive controls. Logistic regression was used to evaluate the association of predicted AI content with submission year and abstract characteristics, and adjusted odds ratios (aORs) were computed. RESULTS: Fifteen thousand five hundred and fifty-three abstracts met inclusion criteria. Across detectors, abstracts submitted in 2023 were significantly more likely to contain AI content than those in 2021 (aOR range from 1.79 with Originality to 2.37 with Sapling). Online-only publication and lack of clinical trial number were consistently associated with AI content. With optimal thresholds, 99.5%, 96%, and 97% of GPT-3/4–generated abstracts were identified by GPTZero, Originality, and Sapling respectively, and no sampled abstracts from before 2020 were classified as AI generated by the GPTZero and Originality detectors. Correlation between detectors was low to moderate, with Spearman correlation coefficient ranging from 0.14 for Originality and Sapling to 0.47 for Sapling and GPTZero. CONCLUSION: There is an increasing signal of AI content in ASCO abstracts, coinciding with the growing popularity of generative AI models. Analysis of 15,553 ASCO meeting abstracts with 3 AI detectors suggests rise in AI content in 2023. [ABSTRACT FROM AUTHOR]

This article examines the problem of fabricated citations in AI-generated writing, focusing on large language models like ChatGPT. The author argues that these fake citations reveal issues with both the processing of natural language training data and writing in general. The implications of these false citations for the future of writing and the ethics of accreditation are explored, as well as the limitations and potential dangers of ChatGPT. The article discusses concerns in various fields, such as finance and medicine, and addresses the debate surrounding AI-generated text in scholarly manuscripts and the legal responsibilities of authors. It concludes by discussing the nature of language models and their relationship to human language, suggesting that they are fulfilling poststructuralist predictions about the future of literature. The article also explores the use of language models, specifically the LLM, in generating text and the issues that arise from it. It introduces the concept of "hallucination" to describe the inaccuracies and fabricated citations produced by AI models, while also discussing the controversy surrounding this term and proposing alternative designations. The challenges posed by AI-generated text and their implications for authorship, plagiarism, and privacy are highlighted, emphasizing the need for ethical considerations and a deeper understanding of the role of AI in textual technologies. [Extracted from the article]

In addressing the critical role of emotional context in patient–clinician conversations, this study conducted a comprehensive sentiment analysis using BERT, RoBERTa, GPT-2, and XLNet. Our dataset includes 185 h of Greek conversations focused on hematologic malignancies. The methodology involved data collection, data annotation, model training, and performance evaluation using metrics such as accuracy, precision, recall, F1-score, and specificity. BERT outperformed the other methods across all sentiment categories, demonstrating its effectiveness in capturing the emotional context in clinical interactions. RoBERTa showed a strong performance, particularly in identifying neutral sentiments. GPT-2 showed promising results in neutral sentiments but exhibited a lower precision and recall for negatives. XLNet showed a moderate performance, with variations across categories. Overall, our findings highlight the complexities of sentiment analysis in clinical contexts, especially in underrepresented languages like Greek. These insights highlight the potential of advanced deep-learning models in enhancing communication and patient care in healthcare settings. The integration of sentiment analysis in healthcare could provide insights into the emotional states of patients, resulting in more effective and empathetic patient support. Our study aims to address the gap and limitations of sentiment analysis in a Greek clinical context, an area where resources are scarce and its application remains underexplored. [ABSTRACT FROM AUTHOR]

Comprehensive systematic reviews require many hours of manual review by researchers. This study retrospectively investigates the use of GPT-3.5 Turbo to screen titles and abstracts as a second reviewer or to prescreen citations for human review. The authors describe the process of prompt development as well as the performance of GPT-3.5 Turbo in terms of citations missed and the added number of citations for reconciliation for 5 different already completed reviews. Background: Systematic reviews are performed manually despite the exponential growth of scientific literature. Objective: To investigate the sensitivity and specificity of GPT-3.5 Turbo, from OpenAI, as a single reviewer, for title and abstract screening in systematic reviews. Design: Diagnostic test accuracy study. Setting: Unannotated bibliographic databases from 5 systematic reviews representing 22 665 citations. Participants: None. Measurements: A generic prompt framework to instruct GPT to perform title and abstract screening was designed. The output of the model was compared with decisions from authors under 2 rules. The first rule balanced sensitivity and specificity, for example, to act as a second reviewer. The second rule optimized sensitivity, for example, to reduce the number of citations to be manually screened. Results: Under the balanced rule, sensitivities ranged from 81.1% to 96.5% and specificities ranged from 25.8% to 80.4%. Across all reviews, GPT identified 7 of 708 citations (1%) missed by humans that should have been included after full-text screening at the cost of 10 279 of 22 665 false-positive recommendations (45.3%) that would require reconciliation during the screening process. Under the sensitive rule, sensitivities ranged from 94.6% to 99.8% and specificities ranged from 2.2% to 46.6%. Limiting manual screening to citations not ruled out by GPT could reduce the number of citations to screen from 127 of 6334 (2%) to 1851 of 4077 (45.4%), at the cost of missing from 0 to 1 of 26 citations (3.8%) at the full-text level. Limitations: Time needed to fine-tune prompt. Retrospective nature of the study, convenient sample of 5 systematic reviews, and GPT performance sensitive to prompt development and time. Conclusion: The GPT-3.5 Turbo model may be used as a second reviewer for title and abstract screening, at the cost of additional work to reconcile added false positives. It also showed potential to reduce the number of citations before screening by humans, at the cost of missing some citations at the full-text level. Primary Funding Source: None. [ABSTRACT FROM AUTHOR]

Thirty years ago, we proposed the similarity between the functioning of artificial intelligence and the human psyche, suggesting multiple parallels between the Freudian model proposed in the "Project for Psychology for Neurologists" and the connectionist theories applied in the generation of parallel distributed processing systems (PDP), also known as connectionist models. These models have been and continue to be the foundation of general artificial intelligences like ChatGPT, evolving and gaining prominence in everyday life. From the earliest applications in psychiatry, recreating computationally simulated modes of illnesses, to the use of deep learning models, especially in the field of computer vision for tasks such as image recognition, segmentation, and classification. Recurrent Neural Networks (RNN) and Long Short-Term Memory (LSTM) are employed for tasks involving sequences of data, such as natural language processing, or models based on the Transformer architecture, like BERT and GPT (Generative Pre-trained Transformer), which have revolutionized natural language processing. In this present work, we analyze the significance of the emergence and exponential growth of these types of tools in the field of healthcare, from medical diagnosis and patient care to psychological attention and psychotherapeutic treatment, exploring the changes and transformations in the forms of subjective expression that are arising. We also examine and argue for the importance and validity of the relational dimension proposed by our psychoanalytic approach in contrast to the potential use of these tools as treatment models. [ABSTRACT FROM AUTHOR]

Aim: Large language models (LLMs) have been suggested to play a role in medical education and medical practice. However, the potential of their application in the psychiatric domain has not been well‐studied. Method: In the first step, we compared the performance of ChatGPT GPT‐4, Bard, and Llama‐2 in the 2022 Taiwan Psychiatric Licensing Examination conducted in traditional Mandarin. In the second step, we compared the scores of these three LLMs with those of 24 experienced psychiatrists in 10 advanced clinical scenario questions designed for psychiatric differential diagnosis. Result: Only GPT‐4 passed the 2022 Taiwan Psychiatric Licensing Examination (scoring 69 and ≥ 60 being considered a passing grade), while Bard scored 36 and Llama‐2 scored 25. GPT‐4 outperformed Bard and Llama‐2, especially in the areas of 'Pathophysiology & Epidemiology' (χ2 = 22.4, P < 0.001) and 'Psychopharmacology & Other therapies' (χ2 = 15.8, P < 0.001). In the differential diagnosis, the mean score of the 24 experienced psychiatrists (mean 6.1, standard deviation 1.9) was higher than that of GPT‐4 (5), Bard (3), and Llama‐2 (1). Conclusion: Compared to Bard and Llama‐2, GPT‐4 demonstrated superior abilities in identifying psychiatric symptoms and making clinical judgments. Besides, GPT‐4's ability for differential diagnosis closely approached that of the experienced psychiatrists. GPT‐4 revealed a promising potential as a valuable tool in psychiatric practice among the three LLMs. [ABSTRACT FROM AUTHOR]

Purpose: ChatGPT (Chat-Generative Pre-trained Transformer) has proven to be a powerful information tool on various topics, including healthcare. This system is based on information obtained on the Internet, but this information is not always reliable. Currently, few studies analyze the validity of these responses in rhinology. Our work aims to assess the quality and reliability of the information provided by AI regarding the main rhinological pathologies. Methods: We asked to the default ChatGPT version (GPT-3.5) 65 questions about the most prevalent pathologies in rhinology. The focus was learning about the causes, risk factors, treatments, prognosis, and outcomes. We use the Discern questionnaire and a hexagonal radar schema to evaluate the quality of the information. We use Fleiss's kappa statistical analysis to determine the consistency of agreement between different observers. Results: The overall evaluation of the Discern questionnaire resulted in a score of 4.05 (± 0.6). The results in the Reliability section are worse, with an average score of 3.18. (± 1.77). This score is affected by the responses to questions about the source of the information provided. The average score for the Quality section was 3.59 (± 1.18). Fleiss's Kappa shows substantial agreement, with a K of 0.69 (p < 0.001). Conclusion: The ChatGPT answers are accurate and reliable. It generates a simple and understandable description of the pathology for the patient's benefit. Our team considers that ChatGPT could be a useful tool to provide information under prior supervision by a health professional. [ABSTRACT FROM AUTHOR]

Structured science summaries or research contributions using properties or dimensions beyond traditional keywords enhance science findability. Current methods, such as those used by the Open Research Knowledge Graph (ORKG), involve manually curating properties to describe research papers' contributions in a structured manner, but this is labor-intensive and inconsistent among human domain-expert curators. We propose using Large Language Models (LLMs) to automatically suggest these properties. However, it is essential to assess the readiness of LLMs like GPT-3.5, Llama 2, and Mistral for this task before their application. Our study performs a comprehensive comparative analysis between the ORKG's manually curated properties and those generated by the aforementioned state-of-the-art LLMs. We evaluate LLM performance from four unique perspectives: semantic alignment with and deviation from ORKG properties, fine-grained property mapping accuracy, SciNCL embedding-based cosine similarity, and expert surveys comparing manual annotations with LLM outputs. These evaluations occur within a multidisciplinary science setting. Overall, LLMs show potential as recommendation systems for structuring science, but further fine-tuning is recommended to improve their alignment with scientific tasks and mimicry of human expertise. [ABSTRACT FROM AUTHOR]

Thanks to the publication of ChatGPT, Artificial Intelligence is now basically accessible and usable to all internet users. The technology behind it can be used in many chatbots, whereby the chatbots should be trained for the respective area of application. Depending on the application, the chatbot should react differently and thus, for example, also take on and embody personality traits to be able to help and answer people better and more personally. This raises the question of whether ChatGPT-4 is able to embody personality traits. Our study investigated whether ChatGPT-4's personality can be analyzed using personality tests for humans. To test possible approaches to measuring the personality traits of ChatGPT-4, experiments were conducted with two of the most well-known personality tests: the Big Five and Myers–Briggs. The experiments also examine whether and how personality can be changed by user input and what influence this has on the results of the personality tests. [ABSTRACT FROM AUTHOR]