---
dataset_info:
  features:
  - name: en
    dtype: string
  - name: uk
    dtype: string
  - name: hash
    dtype: string
  - name: source
    dtype: string
  - name: xcomet_xxl
    dtype: float64
  - name: wmt_22_cometkiwi_da_xl
    dtype: float64
  - name: wmt23_cometkiwi_da_xl
    dtype: float64
  - name: wmt23_cometkiwi_da_xxl
    dtype: float64
  - name: metricx23
    dtype: float64
  - name: metricx24
    dtype: float64
  - name: adjusted_metricx23
    dtype: float64
  - name: adjusted_metricx24
    dtype: float64
  - name: avg_metrics
    dtype: float64
  - name: predicted_linear_score
    dtype: float64
  - name: predicted_quadratic_score
    dtype: float64
  - name: predicted_logit_score
    dtype: float64
  - name: conversations
    list:
    - name: from
      dtype: string
    - name: value
      dtype: string
  splits:
  - name: train
    num_bytes: 287644138
    num_examples: 472622
  download_size: 160634330
  dataset_size: 287644138
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---


# Dataset Card for Impossible Questions

## Dataset Description

**Dataset Summary**

This dataset is a a subset of https://huggingface.co/datasets/lang-uk/FiftyFiveShades, filtered only high-quality data based on quadratic score and converted to conversational format.


This dataset enables state of the art translation capabilities, achieving 33 BLEU on FLORES and 31 BLEU on WMT for English-Ukrainian translation pair, both for short- and long-context.

<!--[Provide a brief overview of your dataset - what it contains, its purpose, and why it was created. Example: "This dataset contains X examples of Ukrainian text collected from Y sources, designed to support the development of Ukrainian language models."] -->

**Languages**

- Ukrainian (uk)
- English (en)



## Citation

TBD

<!-- 
**BibTeX**

```bibtex
@dataset{dataset_name,
  author = {[Your Name/Organization]},
  title = {[Dataset Name]},
  year = {2025},
  publisher = {HuggingFace},
  url = {https://huggingface.co/datasets/[your-org]/[dataset-name]}
}
```
-->

## Contact

<!-- For questions or feedback, please contact [your contact information] or open an issue on the dataset repository. -->

For questions or feedback, please open an issue on the dataset repository.

## License

TBD

---

*This dataset is part of the "Lapa" - Ukrainian LLM initiative to advance natural language processing for the Ukrainian language.*