Although many researchers use pre-trained models to better solve downstream tasks, further exploration of more effective pre-training methods remains necessary, especially for multi-modal pre-training where high-quality training data is more difficult to obtain. This work aims to improve the knowledge-learning performance in multi-modal pre-training. Some researchers focus on injecting entity knowledge into language pre-trained models based on masked entity model (MEM) training, which masks entities randomly and lets the model recover. These methods cannot guarantee good performance due to the lack of consideration of which entities are more valuable for learning. Moreover, in multi-modal training data, some entities may be unrelated to visual content. In this work, for the vision-language pre-trained model, we propose a Masked Entity Model pre-training method based on Active learning (ActiveMEM). It is designed to actively mask important and informative entities—those that are both informative and uncertain—for the model to recover, thereby encouraging it to extract more valuable knowledge from the data. The proposed method is evaluated using three pre-training datasets and four downstream datasets, and the experimental results demonstrate the effectiveness of our method.

Railway scene understanding is critical for ensuring train operational safety and advancing intelligent railway systems. Existing railway vision methods mainly focus on perception and classification, while lacking regulation-guided semantic reasoning capabilities in complex environments. To address these limitations, this paper proposes a retrieval-augmented generation (RAG)-enhanced vision–language framework for railway signal cognition and safety reasoning. The proposed method integrates railway signal perception, regulatory knowledge retrieval, and multi-modal reasoning to improve factual consistency, reasoning reliability, and operational interpretability. In addition, a dedicated railway signal dataset comprising 500 standardized railway scene images with structured QA annotations is constructed to support regulation-oriented multi-modal recognition evaluation. Experimental results show that the proposed framework improves reasoning accuracy from 28.40% to 67.20% with an average end-to-end inference latency of 11.31 s per sample, and the inference speed can be further improved by adjusting experimental configurations to trade off between efficiency and accuracy, demonstrating the potential of RAG-enhanced architectures as a foundational step toward reliable multi-modal cognition in intelligent railway systems.

Accurate trajectory tracking is fundamental to the autonomous operation of unmanned aerial vehicles (UAVs) in complex tasks. While proximal policy optimization (PPO) has shown strong potential in UAV control, its performance is highly sensitive to hyperparameter configuration, and manual tuning is time-consuming due to complex interparameter coupling. This paper proposes CCO–PPO, a framework integrating the cuckoo catfish optimizer (CCO) with PPO for automatic hyperparameter optimization in UAV trajectory tracking. The problem is formulated as a Markov decision process with a 20-dimensional state space, and the CCO performs offline search over a four-dimensional hyperparameter space. Evaluated across seven test environments covering diverse trajectory geometries, wind disturbances, sensor noise, and large-scale scenarios, CCO–PPO achieves the lowest tracking error in all cases. Performance gains over baseline PPO increase monotonically with task complexity, reaching 18.8% under combined wind disturbance and sensor noise, with statistically significant advantages in 85.7% of pairwise comparisons against baseline PPO, SAC, and TD3. Ablation studies confirm that joint optimization of all four hyperparameters is essential under high-disturbance conditions, and comparisons with Bayesian optimization validate the CCO’s superior cross-seed stability. These results demonstrate that metaheuristic hyperparameter optimization substantially enhances policy robustness in high-disturbance UAV trajectory tracking scenarios.