arXiv Open Access 2024

JBBQ: Japanese Bias Benchmark for Analyzing Social Biases in Large Language Models

Hitomi Yanaka Namgi Han Ryoma Kumon Jie Lu Masashi Takeshita +3 lainnya
Lihat Sumber

Abstrak

With the development of large language models (LLMs), social biases in these LLMs have become a pressing issue. Although there are various benchmarks for social biases across languages, the extent to which Japanese LLMs exhibit social biases has not been fully investigated. In this study, we construct the Japanese Bias Benchmark dataset for Question Answering (JBBQ) based on the English bias benchmark BBQ, with analysis of social biases in Japanese LLMs. The results show that while current open Japanese LLMs with more parameters show improved accuracies on JBBQ, their bias scores increase. In addition, prompts with a warning about social biases and chain-of-thought prompting reduce the effect of biases in model outputs, but there is room for improvement in extracting the correct evidence from contexts in Japanese. Our dataset is available at https://github.com/ynklab/JBBQ_data.

Topik & Kata Kunci

Penulis (8)

H

Hitomi Yanaka

N

Namgi Han

R

Ryoma Kumon

J

Jie Lu

M

Masashi Takeshita

R

Ryo Sekizawa

T

Taisei Kato

H

Hiromi Arai

Format Sitasi

Yanaka, H., Han, N., Kumon, R., Lu, J., Takeshita, M., Sekizawa, R. et al. (2024). JBBQ: Japanese Bias Benchmark for Analyzing Social Biases in Large Language Models. https://arxiv.org/abs/2406.02050

Akses Cepat

Lihat di Sumber
Informasi Jurnal
Tahun Terbit
2024
Bahasa
en
Sumber Database
arXiv
Akses
Open Access ✓