.png){kind=icon}
.png){kind=icon}
Return
Cross-lingual semantic alignment and multi-objective reranking for multilingual abstractive summarization
.png){kind=icon}
PRE
AI
P
J
G
DOI：10.1016/j.knosys.2026.116785.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Cross-lingual alignment improves multilingual semantic stability. • Multi-objective reranking reduces factual hallucinations. • mT5-Large evaluated across five multilingual XL-Sum settings. • Semantic consistency improved from 0.4468 to 0.6710; human ratings corroborate gains. • Factual reranking trades ROUGE overlap for semantic reliability.
Keywords:
Multilingual summarization
Abstractive summarization
Cross-lingual semantic alignment
Factuality reranking
Natural language inference
Semantic consistency
Transformer models
Journal
K
IF:
7.6
Papers:
1.2W
Citations:
4.5W
Organization
No organization information available
Cited Papers