Natural Language Model Research Articles

ChatGPT Fervor: The Empowerments, and Challenges of EST Translation Teaching with Natural Language Models

ChatGPT Fervor: The Empowerments, and Challenges of EST Translation Teaching with Natural Language Models

Improving short text classification with augmented data using GPT-3

Abstract GPT-3 is a large-scale natural language model developed by OpenAI that can perform many different tasks, including topic classification. Although researchers claim that it requires only a small number of in-context examples to learn a task, in practice GPT-3 requires these training examples to be either of exceptional quality or a higher quantity than easily created by hand. To address this issue, this study teaches GPT-3 to classify whether a question is related to data science by augmenting a small training set with additional examples generated by GPT-3 itself. This study compares two augmented classifiers: the Classification Endpoint with an increased training set size and the Completion Endpoint with an augmented prompt optimized using a genetic algorithm. We find that data augmentation significantly increases the accuracy of both classifiers, and that the embedding-based Classification Endpoint achieves the best accuracy of about 76%, compared to human accuracy of 85%. In this way, giving large language models like GPT-3 the ability to propose their own training examples can improve short text classification performance.

Review on ChatGPT: History, Applications and Limitations in Chemistry

The advancement in Artificial intelligence has ushered new opportunities in various aspects of life and fields of study. Among the various AI tools developed, chatGPT has gained considerable attention due to its natural language model and interactive capabilities. It has been recognized as a powerful tool for various applications in chemistry such as physical, organic, inorganic, analytical, industrial, environmental, polymer and food chemistry etc. in this review, we focused on the applications of chatGPT in the above named fields of chemistry. We discussed the evolution of chatGPT from the original GPT model to the current chatGPT model and we also examine the limitations of chatGPT. This review provides a comprehensive summary of the history, potential applications and limitations of chatGPT in the field of chemistry, along with in-depth insight on the future prospect of the technology.

Artificial intelligence: a human and natural language model perspective

Artificial intelligence: a human and natural language model perspective

WITNESSING A PARADIGM SHIFT: ASSESSING THE ROLE OF ARTIFICIAL INTELLIGENCE IN THE DOMAIN OF EDUCATION

This scholarly article explores the profound influence of artificial intelligence (AI) on the field of education, with a particular emphasis on the pivotal role played by ChatGPT, a sophisticated natural language model developed by OpenAI. The research comprehensively examines five key paradigms that are set to shape the trajectory of education in the years to come. These paradigms encompass the implications of AI on educational goals, instructional methods, learning materials and approaches, assessment and evaluation mechanisms, and ultimately, the overall learning outcomes. The paper contends that the integration of AI and ChatGPT in educational settings holds significant promise in enhancing students learning effectiveness, promoting wider access to educational resources, and strengthening the monitoring of educational standards. The study highlights the importance of prioritizing students creativity and critical thinking abilities in education, as these cognitive skills are essential for effectively addressing and resolving diverse challenges related to the environment, resources, economy, politics, and other aspects of future life scenarios. By harnessing AI technologies, particularly ChatGPT, novel possibilities and dynamism have been introduced into traditional pedagogical practices and methodologies, leading to a noteworthy shift towards learner-centric educational approaches. Ultimately, the authors argue that ChatGPT represents a considerable potential avenue for shaping the future of education and serving as a transformative catalyst for driving educational reforms on a broader scale.

Evaluation Method of Electric Vehicle Charging Station Operation Based on Contrastive Learning

This paper aims to address the issue of evaluating the operation of electric vehicle charging stations (EVCSs). Previous studies have commonly employed the method of constructing comprehensive evaluation systems, which greatly relies on manual experience for index selection and weight allocation. To overcome this limitation, this paper proposes an evaluation method based on natural language models for assessing the operation of charging stations. By utilizing the proposed SimCSEBERT model, this study analyzes the operational data, user charging data, and basic information of charging stations to predict the operational status and identify influential factors. Additionally, this study compared the evaluation accuracy and impact factor analysis accuracy of the baseline and the proposed model. The experimental results demonstrate that our model achieves a higher evaluation accuracy (operation evaluation accuracy = 0.9464; impact factor analysis accuracy = 0.9492) and effectively assesses the operation of EVCSs. Compared with traditional evaluation methods, this approach exhibits improved universality and a higher level of intelligence. It provides insights into the operation of EVCSs and user demands, allowing for the resolution of supply–demand contradictions that are caused by power supply constraints and the uneven distribution of charging demands. Furthermore, it offers guidance for more efficient and targeted strategies for the operation of charging stations.

Internet Wizard for Enhancing Open-Domain Question-Answering Chatbot Knowledge Base in Education

Chatbots have gained widespread popularity for their task automation capabilities and consistent availability in various domains, including education. However, their ability to adapt to the continuously evolving and dynamic nature of knowledge is limited. This research investigates the implementation of an internet wizard to enhance the knowledge base of an open-domain question-answering chatbot. The proposed approach leverages search engines, particularly Google, and its features, including feature snippets, knowledge graph, and organic search, in conjunction with data science and natural language models. This mechanism empowers the chatbot to dynamically access the extensive and up-to-date knowledge available on the web, enabling the provision of real time and pertinent answers to user queries sourced from web documents. A pilot study in a higher education context evaluated the chatbot’s mechanism and features, confirming its proficiency in generating responses across a broad range of educational and non-educational topics. Positive feedback and high user satisfaction validate these findings. Notably, the chatbot’s dynamic feature of retrieving related or follow-up questions from search engines significantly enhances student engagement and facilitates exploration of supplementary information beyond the curriculum.

Chat Generative Pre-Trained Transformer (ChatGPT) usage in healthcare

In recent years, with the rapid development of deep learning, artificial intelligence (AI) has gradually become a powerful assistant for humans in various aspects. The maturity and advancement of natural language models led to the birth of Chat Generative Pre-Trained Transformer (ChatGPT), generating great potential for healthcare applications. This review summarized the possible application prospects of ChatGPT in healthcare, assessed the limitations that still exist, and suggested possible improvements. In conclusion, the current version of ChatGPT can provide accurate and practical information for doctors and patients, help healthcare professionals to diagnose and treat, promote medical education, and improve the accuracy and efficiency of the healthcare system, etc. However, ChatGPT still has some limitations, as issues such as the accuracy of the information, privacy of data, ethical and moral sentiments still need to be solved. In the future, scientists still need to make efforts to achieve more accurate and effective medical applications with ChatGPT or other AI model by enhancing model performance, filling the data gap, and promoting the standardization of ethical and copyright issues.

Decoupling Source and Semantic Encoding: An Implementation Study

Despite the remarkable achievements of modern communication systems based on Shannon’s theory, there is still considerable room for exploration in information transmission capacity, and semantic communication technology has emerged as a promising approach in this regard. Nonetheless, the benefits of semantic communication remain elusive, and the absence of a unified system model has hindered practical implementation. In this context, we contend that semantic communication can benefit from data distortion and the incorporation of natural language modeling information, such that source coding with semantic modeling information does not compromise the performance of semantic communication systems. To fortify our stance, a novel Separated Data-Semantic Coding (SDSC) system is proposed, which disentangles the source coding and semantic coding. Furthermore, relevant experiments are conducted to validate the contention and the SDSC system. By illuminating the superiority of semantic communication, the research not only contributes to the advancement of semantic communication technologies but also facilitates the development of more practical communication systems.

Appropriateness and Readability of ChatGPT-4-Generated Responses for Surgical Treatment of Retinal Diseases

To evaluate the appropriateness and readability of the medical knowledge provided by ChatGPT-4, an artificial intelligence-powered conversational search engine, regarding common vitreoretinal surgeries for retinal detachments (RDs), macular holes (MHs), and epiretinal membranes (ERMs). Retrospective cross-sectional study. This study did not involve any human participants. We created lists of common questions about the definition, prevalence, visual impact, diagnostic methods, surgical and nonsurgical treatment options, postoperative information, surgery-related complications, and visual prognosis of RD, MH, and ERM, and asked each question 3 times on the online ChatGPT-4 platform. The data for this cross-sectional study were recorded on April 25, 2023. Two independent retina specialists graded the appropriateness of the responses. Readability was assessed using Readable, an online readability tool. The "appropriateness" and "readability" of the answers generated by ChatGPT-4 bot. Responses were consistently appropriate in 84.6% (33/39), 92% (23/25), and 91.7% (22/24) of the questions related to RD, MH, and ERM, respectively. Answers were inappropriate at least once in 5.1% (2/39), 8% (2/25), and 8.3% (2/24) of the respective questions. The average Flesch Kincaid Grade Level and Flesch Reading Ease Score were 14.1 ± 2.6 and 32.3 ± 10.8 for RD, 14 ± 1.3 and 34.4 ± 7.7 for MH, and 14.8 ± 1.3 and 28.1 ± 7.5 for ERM. These scores indicate that the answers are difficult or very difficult to read for the average lay person and college graduation would be required to understand the material. Most of the answers provided by ChatGPT-4 were consistently appropriate. However, ChatGPT and other natural language models in their current form are not a source of factual information. Improving the credibility and readability of responses, especially in specialized fields, such as medicine, is a critical focus of research. Patients, physicians, and laypersons should be advised of the limitations of these tools for eye- and health-related counseling. Proprietary or commercial disclosure may be found after the references.

Error‐Correction Mechanisms in Language Learning: Modeling Individuals

AbstractSince its first adoption as a computational model for language learning, evidence has accumulated that Rescorla–Wagner error‐correction learning (Rescorla & Wagner, 1972) captures several aspects of language processing. Whereas previous studies have provided general support for the Rescorla–Wagner rule by using it to explain the behavior of participants across a range of tasks, we focus on testing predictions generated by the model in a controlled natural language learning task and model the data at the level of the individual learner. By adjusting the parameters of the model to fit the trial‐by‐trial behavioral choices of participants, rather than fitting a one‐for‐all model using a single set of default parameters, we show that the model accurately captures participants’ choices, time latencies, and levels of response agreement. We also show that gender and working memory capacity affect the extent to which the Rescorla–Wagner model captures language learning.

Automatic Generation of SBML Kinetic Models from Natural Language Texts Using GPT.

Kinetic modeling is an essential tool in systems biology research, enabling the quantitative analysis of biological systems and predicting their behavior. However, the development of kinetic models is a complex and time-consuming process. In this article, we propose a novel approach called KinModGPT, which generates kinetic models directly from natural language text. KinModGPT employs GPT as a natural language interpreter and Tellurium as an SBML generator. We demonstrate the effectiveness of KinModGPT in creating SBML kinetic models from complex natural language descriptions of biochemical reactions. KinModGPT successfully generates valid SBML models from a range of natural language model descriptions of metabolic pathways, protein-protein interaction networks, and heat shock response. This article demonstrates the potential of KinModGPT in kinetic modeling automation.

Early Identification of Family Physicians Using Qualitative Admissions Data.

The medical community has been concerned about the shortage of family physicians for decades. Identification of likely family medicine (FM) student matches early in medical school is an efficient recruitment tool. The objective of this study was to analyze qualitative data from medical school applications to establish themes that differentiate future family physicians from their non-FM counterparts. We conducted a qualitative analysis of admissions essays from two groups of 2010-2019 medical school graduates: a study group of students who matched to FM (n=135) and a random sample comparison group of non-FM matches (n=136). We utilized a natural language modeling platform to recognize semantic patterns in the data. This platform generated keywords for each sample, which then guided a more traditional content analysis of the qualitative data for themes. The two groups shared two themes: emotions and science/academics, but with some differences in thematic emphasis. The study group tended toward more positive emotions and the comparison group tended to utilize more specialized scientific language. The study group exhibited two unique themes: special interests in service and community/people. A secondary theme of religious faith was evident in the FM study group. The comparison group exhibited two unique themes: lab/clinical research and career aspirations. Aided by machine learning, a novel analytical approach revealed key differences between FM and non-FM student application materials. Findings suggest qualitative application data may contain identifiable thematic differences when comparing students who eventually match into FM residency programs to those who match into other specialties. Assessing student potential for FM could help guide recruitment and mentorship activities.

Predicting stock market using natural language processing

PurposePredicting the stock market's prices has always been an interesting topic since its closely related to making money. Recently, the advances in natural language processing (NLP) have opened new perspectives for solving this task. The purpose of this paper is to show a state-of-the-art natural language approach to using language in predicting the stock market.Design/methodology/approachIn this paper, the conventional statistical models for time-series prediction are implemented as a benchmark. Then, for methodological comparison, various state-of-the-art natural language models ranging from the baseline convolutional and recurrent neural network models to the most advanced transformer-based models are developed, implemented and tested.FindingsExperimental results show that there is a correlation between the textual information in the news headlines and stock price prediction. The model based on the GRU (gated recurrent unit) cell with one linear layer, which takes pairs of the historical prices and the sentiment score calculated using transformer-based models, achieved the best result.Originality/valueThis study provides an insight into how to use NLP to improve stock price prediction and shows that there is a correlation between news headlines and stock price prediction.

A new horizon for possibility thinking: A conceptual case study of Human × AI collaboration

Possibility thinking (PT) can be described as an imaginative and action-oriented process that drives movement from what is to what could be. The promise of PT is most fully realized in collaboration with others. With recent advances in AI, it can be argued that natural language models (e.g. OpenAI’s GPT models) represent a possible “other” for such collaborations. This article aims to illustrate, through a conceptual case study, how Human × AI collaboration can support PT. The article opens with a brief overview of the basic principles of PT and how those principles can be applied in Human × AI collaboration. Next, a conceptual case study will be presented, which reports on six examples of Human × AI collaborations used for PT. The article closes with a brief discussion of implications and future directions, including the importance of developing a principled approach when using AI for PT. This article likely will be of interest to broad audiences within and across domains, including readers focused on human-centered AI collaborations and anyone interested in learning more about new and emerging ways to generate possibilities in their learning, work, and lives.

Learning to Read and Write in the Language of Proteins

Learning to Read and Write in the Language of Proteins

A SWOT analysis of ChatGPT: Implications for educational practice and research

ABSTRACT ChatGPT is an AI tool that has sparked debates about its potential implications for education. We used the SWOT analysis framework to outline ChatGPT’s strengths and weaknesses and to discuss its opportunities for and threats to education. The strengths include using a sophisticated natural language model to generate plausible answers, self-improving capability, and providing personalised and real-time responses. As such, ChatGPT can increase access to information, facilitate personalised and complex learning, and decrease teaching workload, thereby making key processes and tasks more efficient. The weaknesses are a lack of deep understanding, difficulty in evaluating the quality of responses, a risk of bias and discrimination, and a lack of higher-order thinking skills. Threats to education include a lack of understanding of the context, threatening academic integrity, perpetuating discrimination in education, democratising plagiarism, and declining high-order cognitive skills. We provide agenda for educational practice and research in times of ChatGPT.

Assessing structure and disorder prediction tools for de novo emerged proteins in the age of machine learning.

Background: De novo protein coding genes emerge from scratch in the non-coding regions of the genome and have, per definition, no homology to other genes. Therefore, their encoded de novo proteins belong to the so-called "dark protein space". So far, only four de novo protein structures have been experimentally approximated. Low homology, presumed high disorder and limited structures result in low confidence structural predictions for de novo proteins in most cases. Here, we look at the most widely used structure and disorder predictors and assess their applicability for de novo emerged proteins. Since AlphaFold2 is based on the generation of multiple sequence alignments and was trained on solved structures of largely conserved and globular proteins, its performance on de novo proteins remains unknown. More recently, natural language models of proteins have been used for alignment-free structure predictions, potentially making them more suitable for de novo proteins than AlphaFold2. Methods: We applied different disorder predictors (IUPred3 short/long, flDPnn) and structure predictors, AlphaFold2 on the one hand and language-based models (Omegafold, ESMfold, RGN2) on the other hand, to four de novo proteins with experimental evidence on structure. We compared the resulting predictions between the different predictors as well as to the existing experimental evidence. Results: Results from IUPred, the most widely used disorder predictor, depend heavily on the choice of parameters and differ significantly from flDPnn which has been found to outperform most other predictors in a comparative assessment study recently. Similarly, different structure predictors yielded varying results and confidence scores for de novo proteins. Conclusions: We suggest that, while in some cases protein language model based approaches might be more accurate than AlphaFold2, the structure prediction of de novo emerged proteins remains a difficult task for any predictor, be it disorder or structure.

Chatgpt4, DALL·E, Bard, Claude, BERT: Open Possibilities

The past year has seen explosive growth in artificial intelligence systems[1] aimed at augmenting human creativity in unprecedented ways. Systems like OpenAI's ChatGPT and DALL-E 2, Google's Bard, Anthropic's Claude, and natural language models like BERT have demonstrated the rapid pace of progress in this field[2, 3]. Their capabilities to generate human-like text, create original images from text descriptions, answer complex questions, summarize documents, and more seemingly expand by the day. These AI tools mark a shift from systems focused purely on optimizing for accuracy towards models adept at more subjective, ambiguous, and open-ended tasks. This more flexible approach better captures the complexity of language, creativity, and cognition. The results are systems with which we can brainstorm ideas, get unbiased feedback, develop stories, design prototypes, and enhance our own creative output.

The potential impact of emerging technologies on democratic representation: Evidence from a field experiment

Advances in machine learning have led to the creation natural language models that can mimic human writing style and substance. Here we investigate the challenge that machine-generated content, such as that produced by the model GPT-3, presents to democratic representation by assessing the extent to which machine-generated content can pass as constituent sentiment. We conduct a field experiment in which we send both handwritten and machine-generated letters (a total of 32,398 emails) to 7132 state legislators. We compare legislative response rates for the human versus machine-generated constituency letters to gauge whether language models can approximate inauthentic constituency voices at scale. Legislators were only slightly less likely to respond to artificial intelligence (AI)-generated content than to human-written emails; the 2% difference in response rate was statistically significant but substantively small. Qualitative evidence sheds light on the potential perils that this technology presents for democratic representation, but also suggests potential techniques that legislators might employ to guard against misuses of language models.