Spanish-BFF-2

Avaliable in Hugging Face https://huggingface.co/datasets/MMG/SpanishBFF2

Dataset Description

Spanish-BFF is the first Spanish AI-generated dictionary using GPT3.

• Paper: Building another Spanish dictionary, this time with GPT-4
• Point of Contact: [email protected] , [email protected]

Dataset Summary

Spanish-BFF contains a total of 76,963 lemmas with its definitions.

These lemmas correspond to nominal, adjetival, verbal and adverbial classes.

Languages

• Spanish (es)

Dataset Structure

Data Instances

"retransmisión= [{"pos_tag": "Nombre femenino",
            "definition": "Acción y efecto de retransmitir un evento, generalmente por televisión o radio.",
            "example": "La retransmisión del partido de fútbol fue vista por millones de personas."}]

Data Fields

{
  pos_tag: str
  definition: str
  example: str
}

Data Splits

Split	Size
train	76,963

Considerations for Using the Data

Social Impact of Dataset

This corpus is the second open-source complete dictionary produced by LLMs. We intend to contribute to a better understanding and development of NLP and promote responsible use.

Biases and Hallucinations

This version has not been postprocessed to mitigate potential errors, biases or hallucinations the AI model could have generated.

Citation

@misc{ortegamartín2024building,
      title={Building another Spanish dictionary, this time with GPT-4}, 
      author={Miguel Ortega-Martín and Óscar García-Sierra and Alfonso Ardoiz and Juan Carlos Armenteros and Ignacio Garrido and Jorge Álvarez and Camilo Torrón and Iñigo Galdeano and Ignacio Arranz and Oleg Vorontsov and Adrián Alonso},
      year={2024},
      eprint={2406.11218},
      archivePrefix={arXiv},
      primaryClass={id='cs.CL' full_name='Computation and Language' is_active=True alt_name='cmp-lg' in_archive='cs' is_general=False description='Covers natural language processing. Roughly includes material in ACM Subject Class I.2.7. Note that work on artificial languages (programming languages, logics, formal systems) that does not explicitly address natural-language issues broadly construed (natural-language processing, computational linguistics, speech, text retrieval, etc.) is not appropriate for this area.'}
}