@article{ART003332399},
author={Dongju Kim and KOH, SEOK JOO},
title={Inference Comparison of Lightweight LLMs for Korean Spam Detection on Edge Devices},
journal={Journal of Internet of Things and Convergence},
issn={2466-0078},
year={2026},
volume={12},
number={2},
pages={95-101}
TY - JOUR
AU - Dongju Kim
AU - KOH, SEOK JOO
TI - Inference Comparison of Lightweight LLMs for Korean Spam Detection on Edge Devices
JO - Journal of Internet of Things and Convergence
PY - 2026
VL - 12
IS - 2
PB - The Korea Internet of Things Society
SP - 95
EP - 101
SN - 2466-0078
AB - This study aims to identify the optimal lightweight language model (LLM) for real-time Korean spam SMS detection on the Jetson Orin Nano edge platform. The methodology utilizes a legal-category dataset of 148,937 samples and a specialized four-stage Korean preprocessing pipeline—comprising morphological analysis, surface normalization, dictionary-based conversion, and tokenization— established in a prior study. Three lightweight models, Gemma3-1B, TinyLlama-1.1B, and DeepSeek-1.3B, were comparatively evaluated using Macro-F1 and per-category Recall across four legal categories defined under Article 50 of the Act on Promotion of Information and Communications Network Utilization and Information Protection.Experimental results demonstrate that Gemma3-1B achieved a Macro-F1 of 0.927 and an 'Illegal Activity' category Recall of 0.941, effectively minimizing False Negatives for high-risk violations while maintaining a Perplexity increase within 4.3% after INT4 quantization. The performance gap between Gemma3-1B and TinyLlama-1.1B was statistically significant (paired t-test: , Hedges' ). These findings confirm that Gemma3-1B is the most suitable model for Korean spam detection systems, as it balances classification accuracy, legal-category reliability, and edge deployability.
KW - Korean Spam Detection;Lightweight Language Model;Legal-based Classification;Edge Computing;Transfer Learning
DO -
UR -
ER -
Dongju Kim and KOH, SEOK JOO. (2026). Inference Comparison of Lightweight LLMs for Korean Spam Detection on Edge Devices. Journal of Internet of Things and Convergence, 12(2), 95-101.
Dongju Kim and KOH, SEOK JOO. 2026, "Inference Comparison of Lightweight LLMs for Korean Spam Detection on Edge Devices", Journal of Internet of Things and Convergence, vol.12, no.2 pp.95-101.
Dongju Kim, KOH, SEOK JOO "Inference Comparison of Lightweight LLMs for Korean Spam Detection on Edge Devices" Journal of Internet of Things and Convergence 12.2 pp.95-101 (2026) : 95.
Dongju Kim, KOH, SEOK JOO. Inference Comparison of Lightweight LLMs for Korean Spam Detection on Edge Devices. 2026; 12(2), 95-101.
Dongju Kim and KOH, SEOK JOO. "Inference Comparison of Lightweight LLMs for Korean Spam Detection on Edge Devices" Journal of Internet of Things and Convergence 12, no.2 (2026) : 95-101.
Dongju Kim; KOH, SEOK JOO. Inference Comparison of Lightweight LLMs for Korean Spam Detection on Edge Devices. Journal of Internet of Things and Convergence, 12(2), 95-101.
Dongju Kim; KOH, SEOK JOO. Inference Comparison of Lightweight LLMs for Korean Spam Detection on Edge Devices. Journal of Internet of Things and Convergence. 2026; 12(2) 95-101.
Dongju Kim, KOH, SEOK JOO. Inference Comparison of Lightweight LLMs for Korean Spam Detection on Edge Devices. 2026; 12(2), 95-101.
Dongju Kim and KOH, SEOK JOO. "Inference Comparison of Lightweight LLMs for Korean Spam Detection on Edge Devices" Journal of Internet of Things and Convergence 12, no.2 (2026) : 95-101.
1