Abstract: To assist learners to maintain the continuity of online learning and guide the implementation of the optimal learning path, the intelligent tutoring system needs to detect the tendency of learners to withdraw from learning in time and take corresponding intervention measures at the right time. Therefore, online learning session dropout prediction research is necessary. However, compared to traditional course dropouts, session dropouts occurred more frequently and with shorter single study sessions. It requires the accurate prediction of learning session dropout state based on limited learning behavior feature data. Therefore, the fragmentation of learning behavior and the immediacy and accuracy of prediction results are the challenges and difficulties of learning session dropout prediction tasks. For the session dropout prediction task, this paper proposed a unified online learning session dropout prediction model (Uni-LSDPM). Based on the pre-training and fine-tuning paradigm, it uses a multi-layer Transformer structure. In the pre-training stage, a bidirectional attention mechanism is used to learn the representation of sequence of learners’ continuous behavioral interaction features. In the fine-tuning stage, a sequence-to-sequence attentional mechanism is used to learn the sequence combination of the learner’s continuous behavioral interaction features and dropout state. The model is pre-trained and fine-tuned based on the EdNet public dataset, and the best prediction effect is obtained through ablation experiment. Comparative experiments were conducted based on multiple datasets. Experimental results show that Uni-LSDPM outperforms existing models in terms of AUC and ACC, and proves that the model has certain robustness and scalability.