政治情報課題における生成AIの影響に関する一考察 NTCIR-14 QA Lab PoliInfoタスクを事例として

渋木 英潔; 小川 泰弘; 木村 泰知; 乙武 北斗; 内田 ゆず; 高丸 圭一; 門脇 一真; 秋葉 友良; 佐々木 稔; 小林 暁雄

doi:10.11517/pjsai.JSAI2024.0_3M1OS12a05

38th (2024)

Session ID : 3M1-OS-12a-05

DOI https://doi.org/10.11517/pjsai.JSAI2024.0_3M1OS12a05

Conference information

Host: The Japanese Society for Artificial Intelligence

Name : The 38th Annual Conference of the Japanese Society for Artificial Intelligence

Number : 38

Location : [in Japanese]

Date : May 28, 2024 - May 31, 2024

A study on the influence of generative AI on political information issues

NTCIR-14 QA Lab PoliInfo task as an example

*Hideyuki SHIBUKI, Yasuhiro OGAWA, Yasutomo KIMURA, Hokuto OTOTAKE, Yuzu UCHIDA, Keiichi TAKAMARU, Kazuma KADOWAKI, Tomoyoshi AKIBA, Minoru SASAKI, Akio KOBAYASHI

Author information

*Hideyuki SHIBUKI
BESNA Institute Inc.
Yasuhiro OGAWA
Nagoya City University
Yasutomo KIMURA
Otaru University of Commerce
Hokuto OTOTAKE
Fukuoka University
Yuzu UCHIDA
Hokkai-Gakuen University
Keiichi TAKAMARU
Utsunomiya Kyowa University
Kazuma KADOWAKI
The Japan Research Institute, Limited
Tomoyoshi AKIBA
Toyohashi University of Technology
Minoru SASAKI
Ibaraki University
Akio KOBAYASHI
National Agriculture and Food Research Organization

Keywords: NLP, LLM, political information, summarization, QA Lab PoliInfo

CONFERENCE PROCEEDINGS FREE ACCESS

Details

Abstract

With the rise of generative AI such as ChatGPT, natural language processing has changed significantly. We have been holding shared tasks on political information issues since before the rise. In this paper, we present the results of solving the tasks at that time using generative AI, and consider the impact that generative AI has had on solving political information issues.

Corresponding author

Conference information

Register with J-STAGE for free!