发表论文
我们实验室在人工智能教育领域的最新研究成果
本文提出了一个专门针对教育场景的大规模语言模型EduGPT。该模型通过在教育领域数据上的预训练和微调，能够更好地理解和生成教育相关内容，为智能教学提供支持。实验表明，该模型在教育任务上的表现显著优于通用语言模型。
This paper presents a novel approach to knowledge tracing using large language models. We demonstrate that LLMs can effectively model student knowledge states and predict learning outcomes with higher accuracy than traditional methods.
We propose a new multimodal learning framework specifically designed for educational scenarios. Our model can effectively process and integrate text, images, and videos to enhance the learning experience.
本文提出了一种新的模型压缩方法，能够将大规模语言模型有效部署到资源受限的边缘设备上。通过创新的知识蒸馏和量化技术，模型大小减少90%的同时保持了85%以上的性能。
This paper addresses the critical privacy challenges in AI-enabled education systems. We propose a comprehensive framework that ensures student data privacy while maintaining the effectiveness of AI-driven personalized learning.
本文从深度学习的角度重新思考自适应学习系统的设计。我们提出了一种新的自适应机制，能够根据学生的实时反馈动态调整学习内容和难度。
We introduce a novel cross-modal attention mechanism that significantly improves the efficiency of multimodal learning in educational contexts. Our method achieves state-of-the-art performance while reducing computational costs.