相关实验视频

Updated: Sep 14, 2025

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

PEGASUS-XL具有以突出为导向的评分和长输入编码,用于多文档抽象总结

Rawan Alsultan¹, Alaa Sagheer², Hala Hamdoun¹

¹Department of Computer Science, College of Computer Sciences and Information Technology, King Faisal University, Hofuf, Saudi Arabia.

Scientific reports

|July 21, 2025

概括

通过整合突出模型和长输入编码,PEGASUS-XL增强了多文档总结. 这种框架产生了更加连贯和信息化的摘要,优于现有的方法.

科学领域:

自然语言处理自然语言处理.
人工智能的人工智能
信息检索信息检索

背景情况:

数字内容的指数增长需要有效的信息合成方法.
多文档摘要 (MDS) 旨在从多个来源创建连贯的摘要.
现有方法在突出内容选择,冗余减少,事实一致性和输入长度限制方面面临挑战.

研究的目的:

引入PEGASUS-XL,这是MDS的一个增强的抽象总结框架.
解决MDS的关键挑战,包括信息合成和处理长输入.
提高生成的摘要的质量,连贯性和准确性.

主要方法:

开发了一个结构化的增强管道,整合了词汇语义显著性建模和长输入编码.
采用混合评分机制 (TF-IDF,SBERT) 与适应权重用于内容选择.
使用的最大边际相关性 (MMR) 以减少多样性和冗余性.
集成的Longformer克服了输入长度限制和微调的PEGASUS进行抽象的总结.

主要成果:

在Multi-News和XSum数据集上,PEGASUS-XL的表现始终优于强大的基线 (BART,PRIMERA).
在多个评估指标 (ROUGE,METEOR,BERTScore,SBERT相似性) 中取得了卓越的表现.
人类评估证实了摘要的增强连贯性,信息性和准确性.

结论:

在多文档场景中,PEGASUS-XL提供了一个强大,可扩展和可扩展的解决方案,用于高质量的抽象总结.
该框架显示了显著的质量提升,没有显著的计算开销.
进一步的研究可以解决剩余的问题,如事实漂移和剩余冗余.

关键词:

抽象的总结抽象的总结巴特·巴特 (BART BART) 是一个著名的艺术家.长变频器是一个长变频器.最大的边际相关性多个文件的总结.自然语言处理自然语言处理.佩加索斯 (Pegasus) 是一个高山.第一个是初次.嵌入式 SBERT 嵌入式突出度建模的突出度建模国际货币基金组织-国际货币基金组织

更多相关视频

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

相关概念视频

Improving Translational Accuracy

Improving Translational Accuracy

Pre-mRNA Processing: RNA Splicing

Pre-mRNA Processing: RNA Splicing

Additional Subnuclear Structures

Additional Subnuclear Structures

Upsampling

Upsampling

Managing signal sampling rates is essential in digital signal processing to maintain signal integrity. A decimated signal, characterized by a reduced frequency range due to its lower sampling rate, can be upsampled by inserting zeros between each sample. This upsampling process expands the original spectrum and introduces repeated spectral replicas at intervals dictated by the new Nyquist frequency. To refine this zero-inserted sequence, it is passed through a lowpass filter with a cutoff...

Alternative RNA Splicing

Alternative RNA Splicing

Long-term Potentiation

Long-term Potentiation

Long-term potentiation, or LTP, is one of the ways by which synaptic plasticity—changes in the strength of chemical synapses—can occur in the brain. LTP is the process of synaptic strengthening that occurs over time between pre- and postsynaptic neuronal connections. The synaptic strengthening of LTP works in opposition to the synaptic weakening of long-term depression (LTD) and together are the main mechanisms that underlie learning and memory.