The dual-encoder detector for AI-generated go code

INTRODUCTION Everyone knows that the educational environment sets requirements, rules, and procedures for conducting research and starting students. It is the stakeholders of the educational environment (the state, employers, parents, students) who determine how the learning process should be built and what competencies a student should possess after graduation. All stakeholders agree that the educational process must be built on the principle of academic integrity. What is academic integrity? The answer to this question was sought by the participants of the conference. As it turned out, academic integrity is connected not only with issues of prevention of plagiarism in scientific works, but also with teaching at school, presentation of results of scientific works by researches, relations between teacher and student, state regulation of economic, educational and other relations. I hope that our International Conference on Academic integrity in public administration and educational institutions (APAEI) will become an annual venue for discussing about academic integrity. As Chair of the International Conference on Academic integrity in public administration and educational institutions (APAEI) it is my great pleasure and honor to welcome you all to the first edition of our conference! Kyiv-Tallinn, March 26, 2020 Iryna Mihus, APAEI’2020 Chair

This paper explores the academic integrity considerations of students’ use of Artificial Intelligence (AI) tools using Large Language Models (LLMs) such as ChatGPT in formal assessments. We examine the evolution of these tools, and highlight the potential ways that LLMs can support in the education of students in digital writing and beyond, including the teaching of writing and composition, the possibilities of co-creation between humans and AI, supporting EFL learners, and improving Automated Writing Evaluations (AWE). We describe and demonstrate the potential that these tools have in creating original, coherent text that can avoid detection by existing technological methods of detection and trained academic staff alike, demonstrating a major academic integrity concern related to the use of these tools by students. Analysing the various issues related to academic integrity that LLMs raise for both Higher Education Institutions (HEIs) and students, we conclude that it is not the student use of any AI tools that defines whether plagiarism or a breach of academic integrity has occurred, but whether any use is made clear by the student. Deciding whether any particular use of LLMs by students can be defined as academic misconduct is determined by the academic integrity policies of any given HEI, which must be updated to consider how these tools will be used in future educational environments.

This viewpoint article first explores the ethical challenges associated with the future application of large language models (LLMs) in the context of medical education. These challenges include not only ethical concerns related to the development of LLMs, such as artificial intelligence (AI) hallucinations, information bias, privacy and data risks, and deficiencies in terms of transparency and interpretability but also issues concerning the application of LLMs, including deficiencies in emotional intelligence, educational inequities, problems with academic integrity, and questions of responsibility and copyright ownership. This paper then analyzes existing AI-related legal and ethical frameworks and highlights their limitations with regard to the application of LLMs in the context of medical education. To ensure that LLMs are integrated in a responsible and safe manner, the authors recommend the development of a unified ethical framework that is specifically tailored for LLMs in this field. This framework should be based on 8 fundamental principles: quality control and supervision mechanisms; privacy and data protection; transparency and interpretability; fairness and equal treatment; academic integrity and moral norms; accountability and traceability; protection and respect for intellectual property; and the promotion of educational research and innovation. The authors further discuss specific measures that can be taken to implement these principles, thereby laying a solid foundation for the development of a comprehensive and actionable ethical framework. Such a unified ethical framework based on these 8 fundamental principles can provide clear guidance and support for the application of LLMs in the context of medical education. This approach can help establish a balance between technological advancement and ethical safeguards, thereby ensuring that medical education can progress without compromising the principles of fairness, justice, or patient safety and establishing a more equitable, safer, and more efficient environment for medical education.

The rapid evolution of Generative AI (Gen AI) and its integration into higher education necessitates a critical evaluation of teaching and learning practices. Many students already use publicly available AI Digital Assistants (p-AIDA) such as ChatGPT for academic purposes. However, there are concerns around the use of such p-AIDA tools, in particular in terms of data privacy, academic integrity, Intellectual Property (IP), and the impact on the quality of education. We therefore explored the perspectives of AI usage with 106 students in two consecutive studies using a Design-Based Research (DBR) approach. Study 1 explored the lived experiences a demoed prototype of an institutional AI Digital Assistant (i-AIDA) with ten distance learning students. Building on these findings, in Study 2 the preliminary results were validated with a cohort of 96 students using an online survey on whether the development of an i-AIDA might mitigate some of these concerns. By comparing i-AIDA with p-AIDA in two studies our research underscores the potential benefits and limitations of in-house Gen AI tools in enhancing learning experiences without compromising academic integrity, IP or privacy.

ChatGPT has gained significant attention in computer programming education due to its advanced capabilities in assisting coding processes and its growing impact on teaching and learning. Despite rapid technological progress and widespread adoption, further research is required to optimise its integration into programming education. This scoping review used the PRISMA–ScR framework to analyse 59 research articles published between 2022 and 2025. The review identified major research areas related to ChatGPT’s use in programming education, including its role as a programming assistant, automated assessment and feedback, student and educator perceptions, curriculum design and instructional strategies, learning outcomes and performance, ethical and academic integrity considerations and applications across specific programming domains. It also examined methodological approaches, participant demographics and geographical distribution across the included studies. Findings highlight benefits of integrating ChatGPT, including enhanced student engagement, increased accessibility, support for bridging knowledge gaps and assistance with code optimisation. Meanwhile, challenges include risks of overreliance, reduced critical thinking, accuracy limitations and academic integrity concerns. This review provides practical insights for educators, universities, students and researchers. It emphasises using ChatGPT as a learning assistant, implementing clear policies, tailoring artificial intelligence (AI) tools to diverse student needs and guiding future research on effective and ethical AI-driven programming education. Implications for practice or policy: ChatGPT should support debugging, exploration and collaboration rather than code generation. Students must annotate AI outputs, reinforced by oral exams and reflective journals. Educators should blend AI feedback with human evaluation through scaffolded, authentic assessments. Institutions need clear ethical policies, equitable access and staff training. Researchers should use longitudinal, mixed methods studies, while developers design explainable, adaptive and integrity-focused features aligned with course progression.

The rapid growth of Artificial Intelligence (AI) in education has sparked both enthusiasm for its potential and concern about academic integrity and human autonomy. Within higher education, these tensions invite deeper reflection on AI literacy and agency among educators and researchers. This paper explores how two early-career researchers and university teaching staff navigate their agency when engaging with Generative AI (GenAI) tools such as ChatGPT in teaching and research contexts. Using a collaborative autoethnographic approach informed by Vygotsky’s cultural-historical theory, we reflect on our experiences of both utilising and refraining from using GenAI tools across academic practices. Through narrative snapshots, we examine how we balanced the potential of GenAI use for teaching practices and supporting professional growth, against ethical concerns related to integrity, privacy, and intellectual property. The study highlights how educators exercise reflective and situated agency in an AI-mediated landscape, demonstrating how everyday academic practices contribute to responsible, human-centred engagement with AI in education.

There seems to be an awakening in Asia and the Middle East (ME) region in the fields of research and publication. In these regions, scholarship and research have not been keeping up with the technologically developed Western world. In response to this realization, governments are heavily investing in academia and rapid academic expansion and reforms are being made to gain a foothold in the academic and research arena. This has led academicians to clamber to publish, which, in turn, has led to issues of academic integrity and intellectual property. It is with this development in mind that the current paper examines the issue of academic integrity in Asia and the Middle East, presenting the challenges being confronted, and suggesting strategies that can be implemented.

Objective: This study examines legislative gaps in copyright protection for AI-generated content, focusing on authorship, originality, and intellectual property rights. It evaluates current legal frameworks and proposes solutions to align them with sustainable development goals. Theoretical Framework: Grounded in copyright law principles, particularly the Berne Convention and regulations in the EU, China, and the U.S., the study analyzes how different jurisdictions handle AI-generated works. Methodology: Using analytical, comparative, and case study methods, the research assesses legal inconsistencies and real-world implications of AI-created content. Results and Discussion: AI enhances education and creative industries but raises ethical and legal concerns, especially in academic integrity and legal translations. Existing copyright laws lack clarity on AI-generated works, necessitating legal reforms to balance innovation and intellectual property protection. Research Implications: The study calls for revising copyright laws to define originality in AI works, mandate registration of AI-assisted content, and utilize blockchain for protection. It also suggests establishing a specialized IP court in the UAE and requiring insurance policies for AI firms. Originality/Value: By providing a comparative analysis and proposing concrete legal reforms, this research contributes to the ongoing discourse on AI and copyright, ensuring laws evolve with technological advancements.

PurposeThe purpose of this paper is to discuss changes in the roles of librarians against the backdrop of emerging technologies.Design/methodology/approachThrough reflection on practice drawn from the author’s 30 years in librarianship, this paper explores the evolving role of librarians in the constantly developing technological environment and demonstrates how librarians can combine information resources, technology and research assistance in these constantly changing spaces. It also explores how technology enhances the role of the librarian, with specific reference to ChatGPT.FindingsThe author stressed the need for library courses for credit to allow for feedback, assessment and critical thinking. The need for continuing professional development for librarians was highlighted. Additionally, areas such as Scholarly Publishing, Metrics and Analytics, Academic Integrity and Intellectual Property were identified as areas of focus for the 21st century Librarian.Research limitations/implicationsThis study is limited to changes in roles as a result of emerging technologies which impact librarians. It is set in the Caribbean.Practical implicationsThis paper is relevant for librarians, regardless of their geographical location, who are also required to be fluid, keep on the cutting edge, adapt and adopt to deliver service in the face of constantly advancing technology.Originality/valueThis paper is a reflection on an original experience from a Caribbean territory. it also explores recent developments in technology, more specifically ChatGPT, and its impact on the roles of librarians. It provides contributions to the ongoing discourse on the evolving role of the librarian from a different perspective.

This article examines a modern acute problem: the formation of academic integrity of future educational managers, who must not only learn the essence of the current laws of Ukraine regarding the protection of intellectual property and copyright during their studies in domestic specialized institutions of higher education. First of all, they must acquire a reputation, authority, even an aura of a kind of respectability of the image of a teacher-leader (a teacher is also for a team of colleagues, parents, and not only for pupils, students), which is important not only for themselves – as managers of pedagogical affairs, but also for pupils and students. – for your successful career. Therefore, they will have weight for valuable in science, because they are original, their author's possessions (as well as in high art) – material manifestations of individual imagination, so important from the point of view of authorship rights, intellectual property, and at the same time - from the point of view of morality: conscience, nobility , honesty, because all this integrally constitutes the true originality of the creative product, which is submitted to the review of the scientific and social public.
 Therefore, education of such personal traits can form a special kind of value competences of future managers of education – spiritual. And they will undoubtedly, over time, become characteristic not only of the most progressive educational and scientific community of Ukraine, but literally of the entire world.
 But the leitmotif of the article lies not in the utopian plane of ideal ideas about an ideal society, but in practical recommendations on methods of forming academic integrity in future educational managers. Thus, the assimilation by students of a whole series of laws of the current legislation of Ukraine on copyright and related to them, on copyright and protection of intellectual property should gradually, on education in the process of training and practice, be integrated in the minds of future managers, becoming a generator of the movement towards honest work for the good future generations.

Adoption of pre-trained large language models (LLMs) across an increasingly diverse range of tasks and domains poses significant problems for authorial attribution and other basic knowl­edge organization practices. Utilizing methods from value-sensitive design, this paper examines the theoretical, practical, and ethical issues introduced by LLMs and describes how their use challenges the supposedly firm boundaries separating specific works and creators. Focusing on the implications of LLM usage for higher education, we use hypothetical value scenarios and stakeholder analysis to weigh the pedagogical risks and benefits of LLM usage, assessing the consequences of their use on and beyond college campuses. While acknowledging the unique challenges presented by this emerging educational trend, we ultimately argue that the issues associated with these novel tools are indicative of preexisting limitations within standard entity-relationship models, not wholly new issues ushered in by the advent of a relatively young technology. We contend that LLM-generated texts largely exacerbate, rather than invent from scratch, the preexisting faults that have frequently posed problems to those seeking to determine, ascribe, and regulate authorship attributions. As the growing popularity of generative AI raises concerns about plagiarism, academic integrity, and intellectual property, we advocate for a reevaluation of reductive work-creator associations and encourage the adoption of more expansive authorial concepts.

Light-weighted ensemble network with multilevel activation visualization for robust diagnosis of COVID19 pneumonia from large-scale chest radiographic database

Automated code generation has revolutionized software development, enabling developers to accelerate project timelines and reduce manual coding errors significantly. As reliance on these technologies grows, the inherent weaknesses of generated code become increasingly apparent. Recent studies have shown that code produced by AI is not inherently safer or of higher quality than human-written code, often replicating existing vulnerabilities. <p xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">To this end, we propose SECURECODER, which integrates Retrieval-Augmented Generation (RAG) with Common Weakness Enumeration (CWE). SECURECODER first utilizes the advanced reasoning capabilities of large language models (LLMs) to generate natural language descriptions of the code’s core business logic and functionality. Then, from a semantic perspective, it matches the requirements of the code generation task with the CWE descriptions through a multi-label classification process. Finally, based on the matched CWE, SECURECODER generates a list of security guidelines the code generation model must adhere to. Breaking down end-to-end code generation tasks into single-target tasks that LLMs excel at ensures that the generated code not only meets functional requirements but also adheres to best security practices, thereby enhancing the interpretability of the automated code generation process. After evaluating 2 programming languages and 7 LLMs on Coploit-generated code, SECURECODER has great generalization capability and could be applied to more programming languages and vulnerability types. SECURECODER could significantly decrease the security weakness in the AI-generated code and is able to mitigate more than 65% of vulnerabilities exposed to software developers. Compared to the baseline open-source LLMs, code vulnerabilities were reduced by at least 14% and the code business logic was not affected.

Code summarization and generation empower conversion between programming language(PL) and natural language (NL), while code translation avails the migration of legacy code from one PL to another. This paper introduces PLBART, a sequence-to-sequence model capable of performing a broad spectrum of program and language understanding and generation tasks. PLBART is pre-trained on an extensive collection of Java and Python functions and associated NL text via denoising autoencoding. Experiments on code summarization in the English language, code generation, and code translation in seven programming languages show that PLBART outperforms or rivals state-of-the-art models. Moreover, experiments on discriminative tasks,e.g., program repair, clone detection, and vulnerable code detection, demonstrate PLBART’s effectiveness in program understanding. Furthermore, analysis reveals that PLBART learns program syn-tax, style (e.g., identifier naming convention), logical flow (e.g., if block inside an else block is equivalent to else if block) that are crucial to program semantics and thus excels even with limited annotations.

The emergence of Unified Modeling Language (UML) as a standard for modeling systems has encouraged the use of automated software tools that facilitate the development process from analysis through coding. The UML state machine diagram is used for modeling the dynamic aspects of systems and includes many concepts that are not present in most popular programming languages, like Java or C++. There exists a gap between high level modeling language and a programming language. This paper proposes an approach to narrow the gap between multiple UML models and an implemented system. The narrowing of gap is achieved by generating compact and efficient Java code directly from UML state machine having sequential and concurrent substates. The code generation is achieved by creating a mapping between UML behavior and the Java programming language. Our model can work as a basis for automatic code generation for object-oriented systems.