20 July 2023 in IEEE Transactions on Mobile Computing
In the realm of human mobility data analysis, a multitude of constraints result in the publication of sparse, non-uniform implicit trajectories without explicit location information, such as coordinates. Researchers have dedicated substantial efforts towards trajectory recovery, aiming to densify trajectories and gain a more comprehensive understanding of human mobility. However, existing trajectory recovery methods focus on explicit trajectories, and require extensive historical data to capture users’ mobility patterns. Nevertheless, implicit trajectories are usually more sparse than explicit trajectories. Addressing these challenges, we propose TrajBERT, an innovative BERT-based trajectory recovery method with spatial-temporal refinement. TrajBERT employs the Transformer encoder to learn mobility patterns bi-directionally and enhances the predictions by cross-stage temporal refinement. Subsequently, we design an output layer with global spatial refinement with a novel spatial-temporal aware loss function. To evaluate the performance of TrajBERT, we conduct a series of experiments on real-world datasets. Remarkably, TrajBERT yields at least 8.2% performance improvement compared to the state-of-the-art trajectory recovery approachs. Furthermore, TrajBERT successfully mitigates the cold start problem commonly experienced with new users lacking historical trajectories. It also shows superior robustness when faced with extremely sparse trajectories, thus demonstrating its potential as a practical tool in the field of human mobility analysis.
31 July 2025 in ACM Transactions on Software Engineering and Methodology
Trusted Execution Environments (TEEs) are used to safeguard on-device models. However, directly employing TEEs to secure the entire DNN model is challenging due to the limited computational speed. Utilizing GPU can accelerate DNN’s computation speed but widely available commercial GPUs usually lack security protection. To this end, scholars introduce TEE-Shielded DNN Partition (TSDP), a method that protects privacy-sensitive weights within TEEs and offloads insensitive weights to GPUs. Nevertheless, current methods do not consider the presence of a knowledgeable adversary who can access abundant publicly available pre-trained models and datasets. This article investigates the security of the existing methods against such a knowledgeable adversary and reveals their inability to fulfill their security promises. Consequently, we introduce a novel partition before training strategy, which effectively separates privacy-sensitive weights from other components of the model. Our evaluation demonstrates that our approach can offer full model protection with a computational cost reduced by a factor of 10. In addition to traditional CNN models, we also demonstrate the scalability to large language models. Our approach can compress the private functionalities of the large language model to lightweight slices and achieve the same level of protection as the shielding-whole-model baseline.