Beyond Distribution: Investigating Language Models' Understanding of Sino-Korean Morphemes

We investigate whether Transformer-based language models, trained solely on Hangul text, can learn the compositional morphology of Sino-Korean (SK) morphemes, which are fundamental to Korean vocabulary. Using BERT BASE and fastText, we conduct controlled experiments with target words and their "real" vs. "fake" neighbors-pairs that share a Hangul syllable representing the same SK morpheme vs. those that share only the Hangul syllable. Our results show that while both models -especially BERT-distinguish real and fake pairs to some extent, their performance is primarily driven by the frequency of each experimental word rather than a true understanding of SK morphemes. These findings highlight the limits of distributional learning for morpheme-level understanding and emphasize the need for explicit morphological modeling or Hanja-aware strategies to improve semantic representation in Korean language models. Our dataset and analysis code are available at: https://github.com/taeheejeon22/ ko-skmorph-lm.