sfrontull/lld_valbadia-lld_gherd

---

task_categories: - translation language: - it - lld size_categories: - n<1K

Dataset Card: Ladin (Val Badia) - Ladin (Gherdëina)

Overview

Source Paper: "Compensating for Data with Reasoning: Low-Resource Machine Translation with LLMs"

Description:

This dataset consists of parallel sentences in two variants of Ladin, stored in a single Parquet file.

Dataset Structure

• Files:
  • dizionar-lgh-ita.parquet: Contains the Italian - Ladin (Gherdëina) translations.

Format

• File Type: Parquet
• Encoding: UTF-8

Usage

from datasets import load_dataset
data = load_dataset("sfrontull/lld_valbadia-lld_gherd")

Citation

If you use this dataset, please cite the following paper:

@InProceedings{frontull-EtAl:2025:WMT,
  author    = {Frontull, Samuel  and  Ströhle, Thomas  and  Zoli, Carlo  and  Pescosta, Werner  and  Frenademez, Ulrike  and  Ruggeri, Matteo  and  Valentin, Daria  and  Comploj, Karin  and  Perathoner, Gabriel  and  Liotto, Silvia  and  Anvidalfarei, Paolo},
  title     = {Bringing Ladin to FLORES+},
  booktitle      = {Proceedings of the Tenth Conference on Machine Translation (WMT 2025)},
  month          = {November},
  year           = {2025},
  address        = {Suzhou, China},
  publisher      = {Association for Computational Linguistics},
  pages     = {1061--1071},
  abstract  = {Recent advances in neural machine translation (NMT) have opened new possibilities for developing translation systems also for smaller, so-called low-resource, languages. The rise of large language models (LLMs) has further revolutionized machine translation by enabling more flexible and context-aware generation. However, many challenges remain for low-resource languages, and the availability of high-quality, validated test data is essential to support meaningful development, evaluation, and comparison of translation systems. In this work, we present an extension of the FLORES+ dataset for two Ladin variants, Val Badia and Gherdëina, as a submission to the Open Language Data Initiative Shared Task 2025. To complement existing resources, we additionally release two parallel datasets for Gherdëina–Val Badia and Gherdëina–Italian. We validate these datasets by evaluating state-of-the-art LLMs and NMT systems on this test data, both with and without leveraging the newly released parallel data for fine-tuning and prompting. The results highlight the considerable potential for improving translation quality in Ladin, while also underscoring the need for further research and resource development, for which this contribution provides a basis.},
  url       = {https://aclanthology.org/2025.wmt-1.81}
}