ABSTRACT The research on nested named entity recognition (NER) is conducive to providing richer semantic representations and capturing the nested structure among entities, which is crucial for the execution of downstream tasks. This paper aims to summarise the nested NER methods that have been combined with emerging technologies in recent years. We summarise the nested NER methods that are integrated with emerging technologies from three dimensions: model, framework, and data. Additionally, we explore the research progress of nested NER in two scenarios: cross‐lingual modality and multi‐modal in different modalities. Furthermore, we discuss the practical applications of NER technology in five fields: biomedicine, justice, finance, media, and e‐commerce. Through this review, we can clearly see the development trends of nested NER technology under emerging technologies and different modalities, as well as its broad application prospects in various fields. This provides a reference for future exploration directions in nested NER.

In recent years, multimodal large-scale pre-trained models have achieved tremendous success and become a milestone in the field of artificial intelligence, demonstrating the effectiveness of the pre-training and fine-tuning paradigm in the multimodal domain. Thus, multimodal pre-trained models have been widely applied in various fields of daily life, including some privacy-sensitive areas such as medical diagnosis, financial analysis, public safety, and social media management. Fine-tuning multimodal pre-trained models to adapt to specific tasks in these fields often requires first acquiring data from these downstream tasks and then conducting fine-tuning training. During this process, sensitive private information may be inadvertently learned and leaked. Therefore, we propose a privacy-preserving multimodal parameter-efficient fine-tuning framework: Transferrable DP-Adapter Tuning (TDPAT). In this framework, the model owner sends a lightweight adapter and a lossy compression emulator to the data owner, who then fine-tunes the adapter on downstream data with the help of the emulator. During the fine-tuning training of the adapter, DP-SGD (Differentially Private Stochastic Gradient Descent) improved based on the ideas of multimodal contrastive learning, is incorporated to achieve differential privacy protection. The fine-tuned adapter is then returned to the model owner, who plugs it into the entire multimodal pre-trained model. The TDPAT framework simultaneously achieves privacy protection in three aspects: data security, model security, and inference security. Moreover, by utilizing the multimodal parameter-efficient method Adapter, it reduces the fine-tuning costs and improves fine-tuning efficiency while ensuring fine-tuning performance.