Explicit in Document Tagging
Explicit in-document tagging involves automatically assigning labels or tags to elements within a document, aiming to improve information organization, retrieval, and analysis. Current research focuses on leveraging large language models (LLMs) and deep learning architectures like transformers and convolutional neural networks (CNNs) to achieve this, often incorporating techniques like few-shot learning, prompt engineering, and transfer learning to address data scarcity and improve efficiency. This field is significant for diverse applications, ranging from improved knowledge management and enhanced search capabilities in various domains (e.g., music, news, scientific literature) to more efficient processing of speech recognition outputs and automated code annotation.
Papers
Benchmarking zero-shot and few-shot approaches for tokenization, tagging, and dependency parsing of Tagalog text
Angelina Aquino, Franz de Leon
TAG: Boosting Text-VQA via Text-aware Visual Question-answer Generation
Jun Wang, Mingfei Gao, Yuqian Hu, Ramprasaath R. Selvaraju, Chetan Ramaiah, Ran Xu, Joseph F. JaJa, Larry S. Davis