Multi-Modal Language Models as Text-to-Image Model Evaluators

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Jiahui, Ross, Candace, Askari-Hemmat, Reyhane, Sinha, Koustuv, Hall, Melissa, Drozdzal, Michal, Romero-Soriano, Adriana
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909608413495296
author Chen, Jiahui
Ross, Candace
Askari-Hemmat, Reyhane
Sinha, Koustuv
Hall, Melissa
Drozdzal, Michal
Romero-Soriano, Adriana
author_facet Chen, Jiahui
Ross, Candace
Askari-Hemmat, Reyhane
Sinha, Koustuv
Hall, Melissa
Drozdzal, Michal
Romero-Soriano, Adriana
contents The steady improvements of text-to-image (T2I) generative models lead to slow deprecation of automatic evaluation benchmarks that rely on static datasets, motivating researchers to seek alternative ways to evaluate the T2I progress. In this paper, we explore the potential of multi-modal large language models (MLLMs) as evaluator agents that interact with a T2I model, with the objective of assessing prompt-generation consistency and image aesthetics. We present Multimodal Text-to-Image Eval (MT2IE), an evaluation framework that iteratively generates prompts for evaluation, scores generated images and matches T2I evaluation of existing benchmarks with a fraction of the prompts used in existing static benchmarks. Moreover, we show that MT2IE's prompt-generation consistency scores have higher correlation with human judgment than scores previously introduced in the literature. MT2IE generates prompts that are efficient at probing T2I model performance, producing the same relative T2I model rankings as existing benchmarks while using only 1/80th the number of prompts for evaluation.
format Preprint
id arxiv_https___arxiv_org_abs_2505_00759
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Multi-Modal Language Models as Text-to-Image Model Evaluators
Chen, Jiahui
Ross, Candace
Askari-Hemmat, Reyhane
Sinha, Koustuv
Hall, Melissa
Drozdzal, Michal
Romero-Soriano, Adriana
Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
The steady improvements of text-to-image (T2I) generative models lead to slow deprecation of automatic evaluation benchmarks that rely on static datasets, motivating researchers to seek alternative ways to evaluate the T2I progress. In this paper, we explore the potential of multi-modal large language models (MLLMs) as evaluator agents that interact with a T2I model, with the objective of assessing prompt-generation consistency and image aesthetics. We present Multimodal Text-to-Image Eval (MT2IE), an evaluation framework that iteratively generates prompts for evaluation, scores generated images and matches T2I evaluation of existing benchmarks with a fraction of the prompts used in existing static benchmarks. Moreover, we show that MT2IE's prompt-generation consistency scores have higher correlation with human judgment than scores previously introduced in the literature. MT2IE generates prompts that are efficient at probing T2I model performance, producing the same relative T2I model rankings as existing benchmarks while using only 1/80th the number of prompts for evaluation.
title Multi-Modal Language Models as Text-to-Image Model Evaluators
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2505.00759