Generative artificial intelligence (GenAI), achieving human-like capabilities in interpreting, summarising, creating, and predicting language, has sparked significant interest, leading to extensive exploration and discussion in educational applications. However, the frontline practice of education stakeholders or the conceptual discussion of theorists alone is not sufficient to deeply understand and reshape the application of GenAI in education, and rigorous empirical research and data-based evidence are essential. In the past two years, a large number of empirical studies on GenAI in education have emerged, but there is still a lack of systematic reviews to summarise and analyse the current empirical studies in this field to evaluate existing progress and inform future research. Therefore, this work systematically reviews and analyses 48 recent empirical studies on GenAI in education, detailing their general characteristics and empirical findings regarding promises and concerns, while also outlining current needs and future directions. Our findings highlight GenAI’s role as an assistant and facilitator in learning support, a subject expert and instructional designer in teaching support, and its contributions to diverse feedback methods and emerging assessment opportunities. The empirical studies also raise concerns such as the impact of GenAI imperfections on feedback quality, ethical dilemmas in complex task applications, and mismatches between artificial intelligence (AI)-enabled teaching and user competencies. Our review also summarises and elaborates on essential areas such as AI literacy and integration, the impact of GenAI on the efficiency of educational processes, collaborative dynamics between AI and teachers, the importance of addressing students’ metacognition with GenAI, and the potential for transformative assessments. These insights provide valuable guidelines for future empirical research on GenAI in education.

Large language models (LLMs) have emerged as powerful tools in natural language processing (NLP), showing a promising future of artificial generated intelligence (AGI). Despite their notable performance in the general domain, LLMs have remained suboptimal in the field of education, owing to the unique challenges presented by this domain, such as the need for more specialized knowledge, the requirement for personalized learning experiences, and the necessity for concise explanations of complex concepts. To address these issues, this paper presents a novel LLM for education named WisdomBot, which combines the power of LLMs with educational theories, enabling their seamless integration into educational contexts. To be specific, we harness self-instructed knowledge concepts and instructions under the guidance of Bloom’s Taxonomy as training data. To further enhance the accuracy and professionalism of model’s response on factual questions, we introduce two key enhancements during inference, i.e., local knowledge base retrieval augmentation and search engine retrieval augmentation during inference. We substantiate the effectiveness of our approach by applying it to several Chinese LLMs, thereby showcasing that the fine-tuned models can generate more reliable and professional responses.

This paper proposes a novel approach to use artificial intelligence (AI), particularly large language models (LLMs) and other foundation models (FMs) in an educational environment. It emphasizes the integration of teams of teachable and self-learning LLMs agents that use neuro-symbolic cognitive architecture (NSCA) to provide dynamic personalized support to learners and educators within self-improving adaptive instructional systems (SIAIS). These systems host these agents and support dynamic sessions of engagement workflow. We have developed the never ending open learning adaptive framework (NEOLAF), an LLM-based neuro-symbolic architecture for self-learning AI agents, and the open learning adaptive framework (OLAF), the underlying platform to host the agents, manage agent sessions, and support agent workflows and integration. The NEOLAF and OLAF serve as concrete examples to illustrate the advanced AI implementation approach. We also discuss our proof of concept testing of the NEOLAF agent to develop math problem-solving capabilities and the evaluation test for deployed interactive agent in the learning environment.

After the Overall Plan for Deepening the Reform of Education Evaluation in the New Era has been released for over two years, the reform of education evaluation has achieved a good start and important phased outcomes. Promoting the digital transformation of education evaluation and developing Big Data-based education evaluation are the main measures of current evaluation reform. Based on the case study of the Minzu University of China, this paper systematically sorts out the relevant research, constructs the factor model and process model of Big Data-based education evaluation from the perspectives of factors and process of evaluation, puts forward the application idea of Big Data-based education evaluation from the perspectives of full business, full process and full factors, and puts forward the practical path of Big Data-based education evaluation from the aspects of application traction, teacher training and safe operation.