T2I-ReasonBench: Benchmarking Reasoning-Informed Text-to-Image Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sun, Kaiyue, Fang, Rongyao, Duan, Chengqi, Liu, Xian, Liu, Xihui
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914002381045760
author Sun, Kaiyue
Fang, Rongyao
Duan, Chengqi
Liu, Xian
Liu, Xihui
author_facet Sun, Kaiyue
Fang, Rongyao
Duan, Chengqi
Liu, Xian
Liu, Xihui
contents We propose T2I-ReasonBench, a benchmark evaluating reasoning capabilities of text-to-image (T2I) models. It consists of four dimensions: Idiom Interpretation, Textual Image Design, Entity-Reasoning and Scientific-Reasoning. We propose a two-stage evaluation protocol to assess the reasoning accuracy and image quality. We benchmark various T2I generation models, and provide comprehensive analysis on their performances.
format Preprint
id arxiv_https___arxiv_org_abs_2508_17472
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle T2I-ReasonBench: Benchmarking Reasoning-Informed Text-to-Image Generation
Sun, Kaiyue
Fang, Rongyao
Duan, Chengqi
Liu, Xian
Liu, Xihui
Computer Vision and Pattern Recognition
We propose T2I-ReasonBench, a benchmark evaluating reasoning capabilities of text-to-image (T2I) models. It consists of four dimensions: Idiom Interpretation, Textual Image Design, Entity-Reasoning and Scientific-Reasoning. We propose a two-stage evaluation protocol to assess the reasoning accuracy and image quality. We benchmark various T2I generation models, and provide comprehensive analysis on their performances.
title T2I-ReasonBench: Benchmarking Reasoning-Informed Text-to-Image Generation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2508.17472