README.md

---
pretty_name: FinBen · MultiFin (English)
license: other
license_name: public-source
language:
- en
task_categories:
- text-classification
size_categories:
- n<1K
source_datasets:
- extended
tags:
- finance
- benchmark
- thefinai
- finben
- flare
dataset_info:
  features:
  - name: id
    dtype: string
  - name: query
    dtype: string
  - name: answer
    dtype: string
  - name: text
    dtype: string
  - name: choices
    sequence: string
  - name: gold
    dtype: int64
  splits:
  - name: test
    num_bytes: 401324
    num_examples: 546
  download_size: 61270
  dataset_size: 401324
extra_gated_heading: Request access to FinBen · MultiFin (English)
extra_gated_description: This FinBen 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 FinBen 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
---

# FinBen · MultiFin (English)

📄 [Paper](https://arxiv.org/abs/2402.12659) · 💻 [Code](https://github.com/The-FinAI/FinBen) · 🏆 [Leaderboard](https://huggingface.co/spaces/TheFinAI/Open-FinLLM-Leaderboard) · 🌐 [The Fin AI](https://thefin.ai)

> Part of **FinBen** — *FinBen: A Holistic Financial Benchmark for Large Language Models* ([arXiv:2402.12659](https://arxiv.org/abs/2402.12659)).

| | |
|---|---|
| Task | multi-class classification |
| Original dataset | MultiFin (Jørgensen et al., 2023) |
| Evaluation metric | F1, Accuracy |
| Source license | Public |
| Language | en |

## Quick Start

```python
from datasets import load_dataset

ds = load_dataset("TheFinAI/en-multifin", split="test")
print(ds[0])
```

## Example prompt

```text
In this task, you're working with English headlines from the MULTIFIN dataset. This dataset is made up of real-world article headlines from a large accounting firm's websites. Your objective is to categorize each headline according to its primary topic. The potential categories are 'Finance', 'Technology', 'Tax & Accounting', 'Business & Management', 'Government & Controls', and 'Industry'. Your r…
```

## Dataset Structure

| Split | Rows |
|---|---|
| `test` | 546 |

| Field | Description |
|---|---|
| `id` | Example id |
| `query` | Full instruction prompt given to the model |
| `answer` | Gold answer / label text |
| `text` | Raw input text (without instruction) |
| `choices` | Label space |
| `gold` | Index of the gold label in `choices` |

## License

The paper lists the original data as **publicly available** without a specific license (FinBen paper, Table 2); refer to the original source for its terms.

## Citation

Please cite FinBen and the original dataset (MultiFin (Jørgensen et al., 2023)):

```bibtex
@misc{xie2024finbenholisticfinancialbenchmark,
      title={FinBen: A Holistic Financial Benchmark for Large Language Models},
      author={Qianqian Xie and Weiguang Han and Zhengyu Chen and Ruoyu Xiang and Xiao Zhang and Yueru He and Mengxi Xiao and Dong Li and Yongfu Dai and Duanyu Feng and Yijing Xu and Haoqiang Kang and Ziyan Kuang and Chenhan Yuan and Kailai Yang and Zheheng Luo and Tianlin Zhang and Zhiwei Liu and Guojun Xiong and Zhiyang Deng and Yuechen Jiang and Zhiyuan Yao and Haohang Li and Yangyang Yu and Gang Hu and Jiajia Huang and Xiao-Yang Liu and Alejandro Lopez-Lira and Benyou Wang and Yanzhao Lai and Hao Wang and Min Peng and Sophia Ananiadou and Jimin Huang},
      year={2024},
      eprint={2402.12659},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2402.12659},
}
```