The field of conversational AI for education is moving towards a more nuanced understanding of the importance of dialogue dynamics and pedagogical strategies in educational settings. Researchers are exploring the use of dialogue analysis and machine learning techniques to identify effective pedagogical strategies and evaluate the performance of AI-based educational agents. The integration of conversational AI agents into educational settings is also being investigated, with a focus on fostering reflective learning and improving student engagement. Noteworthy papers include: Towards AI Agents for Course Instruction in Higher Education, which presents an innovative approach to designing and evaluating AI-based educational agents, and Teacher Demonstrations in a BabyLM's Zone of Proximal Development for Contingent Multi-Turn Interaction, which introduces a novel framework for improving multi-turn contingency in conversational AI models.