HyFit: Hybrid Fine-Tuning With Diverse Sampling for Abstractive Summarization

Abstractive summarization has made significant progress in recent years, which aims to generate a concise and coherent summary that contains the most important facts from the source document. Current fine-tuning approaches based on pre-training models typically rely on autoregressive and maximum lik...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	IEEE transactions on big data 2024, p.1-12
Hauptverfasser:	Zhao, Shu, Cheng, Yuanfang, Zhang, Yanping, Chen, Jie, Duan, Zhen, Sun, Yang, Wang, Xinyuan
Format:	Artikel
Sprache:	eng
Schlagworte:	Abstractive summarization contrastive learning Decoding hybrid fine-tuning Reinforcement learning sampling method Sampling methods Self-supervised learning Semantics Task analysis Training
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!