Unggah-Ungguh-Javanese-Bert-Classifier

Unggah-Ungguh-Javanese-Bert-Classifier is part of the Unggah-Ungguh's model family, a classifier model for Javanese Honorific Classification task that was mentioned in "Do Language Models Understand Honorific Systems in Javanese?". Check out our paper for more information!

Model description

• Model type: A classifier model trained on a highly curated Unggah-Ungguh dataset that represent Javanese Honorific rules and systems.
• Language(s) NLP: Javanese
• License: CC-BY-NC 4.0
• Finetuned from model: w11wo/javanese-bert-small-imdb

Model Sources

• Project Page: https://javanesehonorifics.github.io/
• Repository: https://github.com/JavaneseHonorifics
• Paper: https://arxiv.org/abs/2502.20864

Using the model

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch
model_path = "JavaneseHonorifics/Unggah-Ungguh-Javanese-Bert-Classifier"
tokenizer = AutoTokenizer.from_pretrained(model_path)
model = AutoModelForSequenceClassification.from_pretrained(model_path)
INPUT_TEXT = "Mbak Srini mangan pecel ajange pincuk"
tokenized_input = tokenizer([INPUT_TEXT], return_tensors="pt", truncation=True, padding=True)
with torch.no_grad():
    outputs = model(**tokenized_input)
    y_pred = outputs.logits.argmax(-1)
print("Predicted class:", y_pred.item())

License and Use

Unggah-Ungguh is licensed under the CC-BY-NC 4.0

Citation

@article{farhansyah2025language,
  title={Do Language Models Understand Honorific Systems in Javanese?},
  author={Farhansyah, Mohammad Rifqi and Darmawan, Iwan and Kusumawardhana, Adryan and Winata, Genta Indra and Aji, Alham Fikri and Wijaya, Derry Tanti},
  journal={arXiv preprint arXiv:2502.20864},
  year={2025}
}