Reproduce “khammoon: From a Linguistic Guideline to a Thai Morpheme Segmenter” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Warit Sirichotedumrong, Adisai Na-Thalang, Chanakan Wittayasakpan, Kritsadha Phatcharoen, Kunat Pipatanakul
The authors introduce khammoon, a Thai morpheme segmenter based on a defined linguistic unit and ordered rules. It learns from corrected large-scale LLM labels using a character-level BiLSTM. On 2,000 WangchanThaiInstruct documents, it achieves boundary precision 0.992, recall 0.993, and F1 0.992, 0.031 above the prompted LLM baseline; evaluation is limited to instruction data, with prior exposure disclosed.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.