LLM-Free Image Captioning Evaluation in Reference-Flexible Settings

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hirano, Shinnosuke, Wada, Yuiga, Matsuda, Kazuki, Otsuki, Seitaro, Sugiura, Komei
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917169594368000
author Hirano, Shinnosuke
Wada, Yuiga
Matsuda, Kazuki
Otsuki, Seitaro
Sugiura, Komei
author_facet Hirano, Shinnosuke
Wada, Yuiga
Matsuda, Kazuki
Otsuki, Seitaro
Sugiura, Komei
contents We focus on the automatic evaluation of image captions in both reference-based and reference-free settings. Existing metrics based on large language models (LLMs) favor their own generations; therefore, the neutrality is in question. Most LLM-free metrics do not suffer from such an issue, whereas they do not always demonstrate high performance. To address these issues, we propose Pearl, an LLM-free supervised metric for image captioning, which is applicable to both reference-based and reference-free settings. We introduce a novel mechanism that learns the representations of image--caption and caption--caption similarities. Furthermore, we construct a human-annotated dataset for image captioning metrics, that comprises approximately 333k human judgments collected from 2,360 annotators across over 75k images. Pearl outperformed other existing LLM-free metrics on the Composite, Flickr8K-Expert, Flickr8K-CF, Nebula, and FOIL datasets in both reference-based and reference-free settings. Our project page is available at https://pearl.kinsta.page/.
format Preprint
id arxiv_https___arxiv_org_abs_2512_21582
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LLM-Free Image Captioning Evaluation in Reference-Flexible Settings
Hirano, Shinnosuke
Wada, Yuiga
Matsuda, Kazuki
Otsuki, Seitaro
Sugiura, Komei
Computer Vision and Pattern Recognition
We focus on the automatic evaluation of image captions in both reference-based and reference-free settings. Existing metrics based on large language models (LLMs) favor their own generations; therefore, the neutrality is in question. Most LLM-free metrics do not suffer from such an issue, whereas they do not always demonstrate high performance. To address these issues, we propose Pearl, an LLM-free supervised metric for image captioning, which is applicable to both reference-based and reference-free settings. We introduce a novel mechanism that learns the representations of image--caption and caption--caption similarities. Furthermore, we construct a human-annotated dataset for image captioning metrics, that comprises approximately 333k human judgments collected from 2,360 annotators across over 75k images. Pearl outperformed other existing LLM-free metrics on the Composite, Flickr8K-Expert, Flickr8K-CF, Nebula, and FOIL datasets in both reference-based and reference-free settings. Our project page is available at https://pearl.kinsta.page/.
title LLM-Free Image Captioning Evaluation in Reference-Flexible Settings
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.21582