Personalized Scientific Figure Caption Generation: An Empirical Study on Author-Specific Writing Style Transfer

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kim, Jaeyoung, Lee, Jongho, Choi, Hongjun, Jang, Sion
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918151497711616
author Kim, Jaeyoung
Lee, Jongho
Choi, Hongjun
Jang, Sion
author_facet Kim, Jaeyoung
Lee, Jongho
Choi, Hongjun
Jang, Sion
contents We study personalized figure caption generation using author profile data from scientific papers. Our experiments demonstrate that rich author profile data, combined with relevant metadata, can significantly improve the personalization performance of multimodal large language models. However, we also reveal a fundamental trade-off between matching author style and maintaining caption quality. Our findings offer valuable insights and future directions for developing practical caption automation systems that balance both objectives. This work was conducted as part of the 3rd SciCap challenge.
format Preprint
id arxiv_https___arxiv_org_abs_2509_25817
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Personalized Scientific Figure Caption Generation: An Empirical Study on Author-Specific Writing Style Transfer
Kim, Jaeyoung
Lee, Jongho
Choi, Hongjun
Jang, Sion
Computation and Language
Computer Vision and Pattern Recognition
We study personalized figure caption generation using author profile data from scientific papers. Our experiments demonstrate that rich author profile data, combined with relevant metadata, can significantly improve the personalization performance of multimodal large language models. However, we also reveal a fundamental trade-off between matching author style and maintaining caption quality. Our findings offer valuable insights and future directions for developing practical caption automation systems that balance both objectives. This work was conducted as part of the 3rd SciCap challenge.
title Personalized Scientific Figure Caption Generation: An Empirical Study on Author-Specific Writing Style Transfer
topic Computation and Language
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.25817