Fudan University recently issued regulations, starting from the three departments enrolled in 20 1 1, master students can graduate without publishing articles in core journals, and doctoral students can publish two articles in core journals (originally three articles), which changed the rule that they must publish articles in core journals to graduate. I want to express my views on my more than ten years of postgraduate training and years of editing experience in Dunhuang studies. My opinion is that, first of all, it is best not to make rigid rules on whether to publish academic papers during school; Secondly, doctoral students and master students should not use uniform standards. Doctoral students can publish academic papers as one of the graduation conditions, and master students should not be required to publish papers. Generally speaking, the research level of master students is not high, and most of the academic papers written are introductory or comprehensive, so it is difficult to enter the academic circle to discuss frontier issues. There are also many papers that are repetitive, and many people publish other people's research results after secondary processing. As a periodical editor, I can't help it. Because, first, many papers have to be screened to see if they have been published; Second, whether the survey results are similar; Third, check whether there is plagiarism in academic opinions and papers. Most human manuscripts come from this level. Some researchers are afraid that they will not publish their papers, so they hang up their tutors' names, some tell their tutors, and many don't, misleading editors, making them worry about drowning a high-level research achievement, but at the same time they are afraid of being entangled and copied. At present, there are different standards for academic journals to become doctoral students, including authoritative journals and core journals. Lanzhou University originally required it to be published in authoritative journals. Take history as an example, that is, social sciences, historical research and archaeological journals in China. Later, studies on the history of China, the history and geography of China's border areas and world religions were added, but the problem was still not solved. It is also stipulated to publish 1 article in the above authoritative journals or 3 articles in core journals. Recently, it was changed to publish 60% of 3 articles or the top 30% of 1 articles in all categories of CSSCI source journals. I don't think this is reasonable. In fact, the right to train doctoral students is virtually given to editors of various magazines. Whether doctoral students can graduate depends on the writing level of dissertations, not the level of publishing papers and magazines. The best way is not to stick to the level, but to submit the academic achievements of doctoral students at school at the same time or before submitting the doctoral thesis, as long as they meet the standards. Even some good doctoral dissertations can graduate without pre-publishing results, and they can still go through strict submission procedures. Talent cultivation is an old topic. The quality of doctoral education depends on the guidance level of tutors, the understanding of doctoral students and the academic atmosphere of the unit. The guidance level of the tutor is the first, and the guidance level is not the research level of the tutor. Many experts are excellent, but they can't guide a decent doctoral student. Of course, an excellent doctoral student can never be guided without a competent tutor. In addition, the scientific laboratory model cannot be used in the field of humanities and social sciences. The relationship between liberal arts tutors and doctoral students can only be the relationship between teachers and students, and it must not be the relationship between bosses and employees in science and engineering, and grades cannot be mixed. Tutor is the guide for doctoral students to enter this field. The tutor will guide the doctoral students to enter the frontier field of the subject in the shortest time, and make a research topic according to the students' research level, ability and knowledge field. Tutors should be good at exploring the potential of doctoral students, teaching by example, encouraging priority and improving students' self-confidence. The savvy and hard work of doctoral students are also the key to training. Without savvy, it is difficult to train doctoral students, and savvy and hard-working doctoral students cannot cultivate high standards. Understanding is the basic condition for doctoral students' innovation, and efforts are the minimum requirement for doctoral students' training. Let them integrate pressure, motivation and confidence, so that they don't need rigid requirements and will produce excellent results. (Author: Institute of Dunhuang Studies, Lanzhou University)