DetailMaster: Can Your Text-to-Image Model Handle Long Prompts?

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Jiao, Qirui, Chen, Daoyuan, Huang, Yilun, Lin, Xika, Shen, Ying, Li, Yaliang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913177181093888
author Jiao, Qirui
Chen, Daoyuan
Huang, Yilun
Lin, Xika
Shen, Ying
Li, Yaliang
author_facet Jiao, Qirui
Chen, Daoyuan
Huang, Yilun
Lin, Xika
Shen, Ying
Li, Yaliang
contents While recent Text-to-Image (T2I) models show impressive capabilities in synthesizing images from brief descriptions, they struggle with the long, detailed prompts required for professional applications. We present DetailMaster, a comprehensive benchmark for evaluating T2I capabilities on long prompts with complex compositional requirements, accompanied by an automated data construction pipeline and an evaluation workflow. Comprising expert-validated prompts averaging 284.89 tokens, our benchmark introduces four critical evaluation dimensions: Character Attributes, Structured Character Locations, Multi-Dimensional Scene Attributes, and Spatial/Interactive Relationships. Evaluations on various general-purpose and long-prompt-optimized models reveal critical performance limitations, showing that weak encoders struggle to preserve syntactic dependencies within prompts and diffusion models suffer from attribute leakage under detail-intensive conditions. Through a controlled ablation study under varying constraints, we further show that high-fidelity generation requires a synergistic combination of expanded prompt limits and long-prompt training. We open-source our dataset and code to foster progress in long-prompt-driven T2I generation.
format Preprint
id arxiv_https___arxiv_org_abs_2505_16915
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DetailMaster: Can Your Text-to-Image Model Handle Long Prompts?
Jiao, Qirui
Chen, Daoyuan
Huang, Yilun
Lin, Xika
Shen, Ying
Li, Yaliang
Computer Vision and Pattern Recognition
Artificial Intelligence
While recent Text-to-Image (T2I) models show impressive capabilities in synthesizing images from brief descriptions, they struggle with the long, detailed prompts required for professional applications. We present DetailMaster, a comprehensive benchmark for evaluating T2I capabilities on long prompts with complex compositional requirements, accompanied by an automated data construction pipeline and an evaluation workflow. Comprising expert-validated prompts averaging 284.89 tokens, our benchmark introduces four critical evaluation dimensions: Character Attributes, Structured Character Locations, Multi-Dimensional Scene Attributes, and Spatial/Interactive Relationships. Evaluations on various general-purpose and long-prompt-optimized models reveal critical performance limitations, showing that weak encoders struggle to preserve syntactic dependencies within prompts and diffusion models suffer from attribute leakage under detail-intensive conditions. Through a controlled ablation study under varying constraints, we further show that high-fidelity generation requires a synergistic combination of expanded prompt limits and long-prompt training. We open-source our dataset and code to foster progress in long-prompt-driven T2I generation.
title DetailMaster: Can Your Text-to-Image Model Handle Long Prompts?
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2505.16915