TY  - JOUR
T1  - Fine-Tuning BGE-M3 for Defense Language EmbeddingModel: The Impact of Negative Sample Selection inContrastive Learning
AU  - Kim, Junsub 
AU  - Choi, Dongnyeok 
AU  - Kim, Sung Gu 
AU  - Kim, Deuk Hwa 
JO  - Journal of KIISE, JOK
PY  - 2026
DA  - 2026/1/14
DO  - 10.5626/JOK.2026.53.2.117
KW  - embedding model
KW  - contrastive learning
KW  - Hard Negative
KW  - BGE-M3
AB  - Korean language models specifically designed for the defense sector are still limited,
even with the rapid advancements in text embeddings. In this study, we fine-tune the multilingual
BGE-M3 model to better understand military terminology and investigate how negative sampling in
contrastive learning impacts downstream performance. We evaluate three strategies: Easy (random
negatives), Hard (lexicographic adjacency), and Harder (similarity-mined negatives). Our analysis,
based on clustering metrics such as Accuracy, NMI, and ARI using a defense news dataset, reveals
that the similarity-based Harder strategy consistently outperforms the others. Further evaluations on
the KorSTS dataset demonstrate that the Harder approach maintains strong Spearman and Pearson
correlations, indicating successful domain adaptation without compromising overall semantic
competence. Interestingly, the three Harder variants—negatives mined with BGE-M3, ko-sroberta, and
multilingual-e5—produce nearly identical similarity distributions and comparable improvements, while
the Easy strategy plateaus and the Hard strategy shows only moderate performance. These findings
suggest that mining sufficiently similar negatives, as opposed to using random or adjacent ones, is
crucial for effective, domain-specific fine-tuning of multilingual embedding models.