README.md

metadata

pretty_name: FLARE-ES · Spanish instruction-tuning data
license: other
license_name: public-source
language:
  - es
task_categories:
  - text-generation
size_categories:
  - 10K<n<100K
source_datasets:
  - extended
tags:
  - finance
  - benchmark
  - thefinai
  - flare-es
  - flare
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: valid
        path: data/valid-*
dataset_info:
  features:
    - name: id
      dtype: string
    - name: conversations
      list:
        - name: from
          dtype: string
        - name: value
          dtype: string
  splits:
    - name: train
      num_bytes: 41354500
      num_examples: 14851
    - name: valid
      num_bytes: 6718150
      num_examples: 2226
  download_size: 23259291
  dataset_size: 48072650
extra_gated_heading: Request access to FLARE-ES · Spanish instruction-tuning data
extra_gated_description: >-
  This FLARE-ES task is released by The Fin AI for research. Access is granted
  automatically after you complete this short form.
extra_gated_button_content: Agree and access
extra_gated_prompt: >-
  By accessing this dataset you agree to the license of the original source
  (Public) and to cite the FLARE-ES paper and the original dataset in any
  resulting publication.
extra_gated_fields:
  Full name: text
  Affiliation: text
  Country: country
  Intended use:
    type: select
    options:
      - Research
      - Education
      - Commercial evaluation
      - Other
  I agree to the terms above and will cite the papers: checkbox

FLARE-ES · Spanish instruction-tuning data

📄 Paper · 🌐 The Fin AI

Part of FLARE-ES — Dólares or Dollars? Unraveling the Bilingual Prowess of Financial LLMs Between Spanish and English (arXiv:2402.07405).

Instruction-tuning conversations used to train the FinMA-ES models; prompts follow the Spanish templates in the paper (Table 3).

Task	instruction tuning (training data, not an evaluation set)
Original dataset	FLARE-ES instruction data built from the Spanish tasks in Table 1
Source license	Public
Language	es

Quick Start

from datasets import load_dataset

ds = load_dataset("TheFinAI/es-instruction-tuning", split="valid")
print(ds[0])

Dataset Structure

Split	Rows
train	14,851
valid	2,226

Field	Description
id	Example id
conversations	Instruction-tuning dialogue turns (from, value)

License

The paper lists the original data as publicly available without a specific license (FLARE-ES paper, Tables 1-2); refer to the original source for its terms.

Citation

Please cite FLARE-ES and the original dataset (FLARE-ES instruction data built from the Spanish tasks in Table 1):

@misc{zhang2024dolaresdollarsunravelingbilingual,
      title={D\'olares or Dollars? Unraveling the Bilingual Prowess of Financial LLMs Between Spanish and English},
      author={Xiao Zhang and Ruoyu Xiang and Chenhan Yuan and Duanyu Feng and Weiguang Han and Alejandro Lopez-Lira and Xiao-Yang Liu and Sophia Ananiadou and Min Peng and Jimin Huang and Qianqian Xie},
      year={2024},
      eprint={2402.07405},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2402.07405},
}