Large language models (LLMs) often show poor performance in low-resource languages like Korean, partly due to unique linguistic challenges such as homophonous Sino-Korean words that are indistinguishable in Hangul script. To address this semantic ambiguity, we propose HanjaBridge, a novel meaning-injection technique integrated into a continual pre-training (CPT) framework. Instead of deterministically mapping a word to a single Hanja (Chinese character), HanjaBridge presents the model with all possible Hanja candidates for a given homograph, encouraging the model to learn contextual disambiguation. This process is paired with token-level knowledge distillation to prevent catastrophic forgetting. Experimental results show that HanjaBridge significantly improves Korean language understanding, achieving a 21% relative improvement on the KoBALT benchmark. Notably, by reinforcing semantic alignment between Korean and Chinese through shared Hanja, we observe a strong positive cross-lingual transfer. Furthermore, these gains persist even when Hanja augmentation is omitted at inference time, ensuring practical efficiency with no additional run-time cost.

The traditional code analysis approach to measure code quality, such as bad smells, coupling, cohesion, complexity, and common weakness enumeration, was a rule-based mechanism. In current, analyzing code through AI is being studied by many researchers. AI code analysis approach trains code patterns to AI model through labeled code datasets. The problem is that AI models require plenty of training datasets for better performance, but there are not enough datasets to train. To solve this problem, we suggest training an AI model with core building blocks of a target language and similar languages together. We can solve two problems the lack of datasets and the low performance of the AI model. Finally, we will compare the performance of the two models. One is a model trained with a dataset containing only one target language, and the other is a model trained with a dataset containing similar languages.