---
license: cc-by-4.0
task_categories:
- conversational
- translation
pretty_name: LigurianNews lij-ita translation dataset (instruction-style)
dataset_info:
  features:
  - name: inputs
    dtype: string
  - name: targets
    dtype: string
  - name: template_id
    dtype: int64
  - name: template_lang
    sequence: string
  splits:
  - name: train
    num_bytes: 288462
    num_examples: 153
  - name: dev
    num_bytes: 47500
    num_examples: 27
  - name: test
    num_bytes: 67307
    num_examples: 36
  download_size: 292727
  dataset_size: 403269
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: dev
    path: data/dev-*
  - split: test
    path: data/test-*
---

This is a Ligurian→Italian document-level translation dataset.
The original data comes from the [LigurianNews corpus](https://github.com/ConseggioLigure/data/), and it has been converted to the instruction format.

The prompts, written in Ligurian, ask the model to "translate the following text to Italian". There are several variants of the prompt which were randomly sampled for each sentence:

```
Traduxi in italian: \<sentence>
Traduxi da-o zeneise à l’italian: \<sentence>
Traduxi da-o ligure à l’italian: \<sentence>
Traduxi sto testo in italian: \<sentence>
Traduxi in lengua italiaña: \<sentence>
Traduxi sto testo da-o zeneise à l’italian: \<sentence>
Traduxi sto testo da-o ligure à l’italian: \<sentence>
Comm’à l’é a traduçion italiaña de sto testo? \<sentence>
Quæ a l’é a traduçion italiaña de sto testo? \<sentence>
Ti peu tradue sto testo in italian? \<sentence>
```

The prompt template used for each dataset entry is referenced in the column `template_id`, with ids ranging from 1 to 10 according to the order given above.
The targets are always prefixed with the string _"A traduçion in italian do testo a l’é: \<sentence>"_ ("The Italian translation of the text is:").

The correspondence between `template_id`, prompt template and target template is therefore:
```
[
    (1, "Traduxi in italian:\n", "A traduçion in italian do testo a l’é:\n"),
    (2, "Traduxi da-o zeneise à l’italian:\n", "A traduçion in italian do testo a l’é:\n"),
    (3, "Traduxi da-o ligure à l’italian:\n", "A traduçion in italian do testo a l’é:\n"),
    (4, "Traduxi sto testo in italian:\n", "A traduçion in italian do testo a l’é:\n"),
    (5, "Traduxi in lengua italiaña:\n", "A traduçion in italian do testo a l’é:\n"),
    (6, "Traduxi sto testo da-o zeneise à l’italian:\n", "A traduçion in italian do testo a l’é:\n"),
    (7, "Traduxi sto testo da-o ligure à l’italian:\n", "A traduçion in italian do testo a l’é:\n"),
    (8, "Comm’à l’é a traduçion italiaña de sto testo?\n", "A traduçion in italian do testo a l’é:\n"),
    (9, "Quæ a l’é a traduçion italiaña de sto testo?\n", "A traduçion in italian do testo a l’é:\n"),
    (10, "Ti peu tradue sto testo in italian?\n", "A traduçion in italian do testo a l’é:\n"),
]
```

The dataset contains 153 train samples, 27 validation samples and 36 test samples.