Optimizing diplomatic indexing: full-parameter vs low-rank adaptation for multi-label classification of diplomatic cables
Computer Science and Information Technologies · 15.12.2025 · 6 (3) · 274-282
DOI: 10.11591/csit.v6i3.pp274-282 ↗Kurz gefasst
Fachartikel in der Zeitschrift Computer Science and Information Technologies, online erschienen am 15.12.2025. Verfasst von Dela Nurlaila und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).
Abstract (Originalsprache)
Accurate classification of diplomatic cables is crucial for Mission’s evaluation and policy formulation. However, these documents often cover multiple topics, hence a multi-label classification approach is necessary. This research explores the application of pre-trained language models (CahyaBERT, IndoBERT, and MBERT) for multi-label classification of diplomatic cable executive summaries, which align with the diplomatic representation index. The study compares full-parameter fine-tuning and low-rank adaptation (LoRA) techniques using cables from 2022-2023. Results demonstrate that Indonesian-specific models, particularly the IndoBERT, outperform multilingual models in classification accuracy. While LoRA showed slightly lower performance than full fine-tuning, it significantly reduced GPU memory usage by 48% and training time by 69.7%. These findings highlight LoRA’s potential for resource-constrained diplomatic institutions, advancing natural language processing in diplomacy and offering pathways for efficient, real-time multi-label classification to enhance diplomatic mission evaluation.
Zitieren
APA
Nurlaila, D., & Girsang, A. S. (2025). Optimizing diplomatic indexing: full-parameter vs low-rank adaptation for multi-label classification of diplomatic cables. Computer Science and Information Technologies, 6(3), 274-282. https://doi.org/10.11591/csit.v6i3.pp274-282
Vancouver
Nurlaila D, Girsang AS. Optimizing diplomatic indexing: full-parameter vs low-rank adaptation for multi-label classification of diplomatic cables. Computer Science and Information Technologies. 2025;6(3):274-282. doi:10.11591/csit.v6i3.pp274-282
BibTeX
@article{Nurlaila2025,
author = {Nurlaila, Dela and Girsang, Abba Suganda},
title = {Optimizing diplomatic indexing: full-parameter vs low-rank adaptation for multi-label classification of diplomatic cables},
journal = {Computer Science and Information Technologies},
year = {2025},
volume = {6},
number = {3},
pages = {274--282},
doi = {10.11591/csit.v6i3.pp274-282}
}
EUVIMED