EUVIMEDEUVIMED entdecken
Crossref

Hybrid rank-MMR

Minh-Tuan Luu, Quoc-Thai Nguyen, Thanh-Huong Le

International journal of electrical and computer engineering systems · 10.11.2026 · 17 (10) · 819-833

DOI: 10.32985/ijeces.17.10.5 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International journal of electrical and computer engineering systems, online erschienen am 10.11.2026. Verfasst von Minh-Tuan Luu und 2 weiteren Autor:innen. Die Arbeit nennt 34 Literaturangaben.

AbstractBibliografische AngabenZitieren

Abstract (Originalsprache)

Text summarization is a challenging problem in natural language processing (NLP), especially for low-resource languages like Vietnamese or specialized documents (such as scientific articles). Currently, extractive text summarization models often rely on semantic matching using pre-trained language models or select sentences based on their position. However, these methods prove ineffective when applied to different languages or domains in extractive text summarization tasks. Therefore, we propose Hybrid Rank-MMR, a re-ranking model that combines semantic similarity with two features: semantic density and sentence position. This method can handle both news (relying on sentence position) and complex datasets (requiring semantic understanding). Furthermore, we refine the MMR algorithm to automatically control diversity and remove duplicate sentences during summary generation. We test on seven datasets in both English (CNN-DailyMail, Multi-News, XScience, WCEP) and Vietnamese (VNDS, ViMs, AbMuSu). The results show that this model significantly improves performance compared to the basic models. ROUGE-1 increases by +2.56 on Multi-News and +6.68 on ViMs. Finally, we integrate an explainable artificial intelligence (XAI) module based on large language models (LLMs). This module uses natural language to explain to the user why the system selected those sentences for summarization. Compared with prior zero-shot extractive systems such as MatchSum, which are applied without fine-tuning, our framework requires no parameter updates on any benchmark.

Zitieren

APA

Luu, M., Nguyen, Q., & Le, T. (2026). Hybrid rank-MMR. International journal of electrical and computer engineering systems, 17(10), 819-833. https://doi.org/10.32985/ijeces.17.10.5

Vancouver

Luu M, Nguyen Q, Le T. Hybrid rank-MMR. International journal of electrical and computer engineering systems. 2026;17(10):819-833. doi:10.32985/ijeces.17.10.5

BibTeX
@article{Luu2026,
  author = {Luu, Minh-Tuan and Nguyen, Quoc-Thai and Le, Thanh-Huong},
  title = {Hybrid rank-MMR},
  journal = {International journal of electrical and computer engineering systems},
  year = {2026},
  volume = {17},
  number = {10},
  pages = {819--833},
  doi = {10.32985/ijeces.17.10.5}
}
Kostenlos starten →