Vietnamese Abstractive Multi Document Summarization

Vietnamese abstractive multi-document summarization (MDS) focuses on automatically generating concise, coherent summaries from multiple Vietnamese news articles on a single topic. Current research emphasizes hybrid approaches combining extractive and abstractive methods, often leveraging pre-trained transformer models like BERT and ViT5, and exploring graph-based techniques to capture inter-sentence relationships within and across documents. This area is significant due to the limited resources for Vietnamese NLP and the potential to improve information access and processing in the Vietnamese language.