To solve a series of problems such as unchanging teaching forms, boring teaching resources, and lack of effective ways to combine teaching and research in traditional classrooms that hinder students’ personalized development and creativity enhancement, an analysis of the innovation of university students’ education management based on the APT model is proposed. This study mainly adopts research methods such as quasi-experimental research method and survey research method and adopts research tools such as test papers, questionnaires and interview outlines to conduct a comprehensive and in-depth analysis and investigation of research data, student research programs, and other works. This study is based on the project “APT Teaching in Smart Classrooms to Promote Deep Learning among University Students”, in which 78 second-year university students majoring in educational technology at University A were selected as research subjects to design a semester-long APT model based on a Rubric and carry out a variety of innovative teaching in smart classrooms. In addition, a quasi-experimental study was conducted to compare the differences in learning performance, research ability, motivation, and attitudes towards the use of the Rubrics between pre- and post-announcement of the Rubrics, providing a new perspective on the differences in the functions and effects of the different types of Rubrics. The results show that APT based on Rubrics improves the academic performance of university students and that the effect of posting Rubrics on activity tasks is better. The Rubric-based APT instruction improves students’ research skills and is more effective when the Rubrics are presented before the activity task.