Personalized Image Caption Generation Using Monte Carlo Tree Search
-
- YOSHIDA Tsukasa
- NTT
-
- SHINBORI Kazuki
- NTT, Meiji University
-
- FUKAYAMA Atsushi
- NTT
Bibliographic Information
- Other Title
-
- モンテカルロ木探索を用いた個人性のある画像キャプション生成
Abstract
<p>This study aims to generate personalized descriptions in image captioning, incorporating individual perspectives and phrasing. With the progress in large language models, achieving notable results in various language tasks is possible. For text generation that reflects individuality, adjusting the language model using limited data from individuals is a challenge. This paper proposes using a personal identification model trained on minimal data combined with Monte Carlo tree search to explore token generation sequences. We demonstrate that this method can produce a broader range of sentences than standard beam search and effectively replicate individuality. </p>
Journal
-
- JSAI Technical Report, SIG-SLUD
-
JSAI Technical Report, SIG-SLUD 100 (0), 01-06, 2024-02-20
The Japanese Society for Artificial Intelligence
- Tweet
Details 詳細情報について
-
- CRID
- 1390862157392167040
-
- ISSN
- 24364576
- 09185682
-
- Text Lang
- ja
-
- Data Source
-
- JaLC
-
- Abstract License Flag
- Allowed