EdiVal-Agent: An Object-Centric Framework for Automated, Fine-Grained Evaluation of Multi-Turn Editing

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Tianyu, Zhang, Yasi, Zhang, Zhi, Yu, Peiyu, Wang, Shu, Wang, Zhendong, Lin, Kevin, Wang, Xiaofei, Yang, Zhengyuan, Li, Linjie, Lin, Chung-Ching, Xie, Jianwen, Leong, Oscar, Wang, Lijuan, Wu, Ying Nian, Zhou, Mingyuan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915870370955264
author Chen, Tianyu
Zhang, Yasi
Zhang, Zhi
Yu, Peiyu
Wang, Shu
Wang, Zhendong
Lin, Kevin
Wang, Xiaofei
Yang, Zhengyuan
Li, Linjie
Lin, Chung-Ching
Xie, Jianwen
Leong, Oscar
Wang, Lijuan
Wu, Ying Nian
Zhou, Mingyuan
author_facet Chen, Tianyu
Zhang, Yasi
Zhang, Zhi
Yu, Peiyu
Wang, Shu
Wang, Zhendong
Lin, Kevin
Wang, Xiaofei
Yang, Zhengyuan
Li, Linjie
Lin, Chung-Ching
Xie, Jianwen
Leong, Oscar
Wang, Lijuan
Wu, Ying Nian
Zhou, Mingyuan
contents Instruction-based image editing has advanced rapidly, yet reliable and interpretable evaluation remains a bottleneck. Current protocols either (i) depend on paired reference images, resulting in limited coverage and inheriting biases from prior generative models or (ii) rely solely on zero-shot vision language models (VLMs), whose prompt-based assessments of instruction following, content consistency, and visual quality are often imprecise. To address this, we introduce EdiVal, an automated and fine-grained evaluation framework grounded in an object-centric perspective, designed to assess not only standard single-turn but also multi-turn instruction-based editing with precision. Given an input image, EdiVal first decomposes it into semantically meaningful objects, then synthesizes diverse, context-aware editing instructions while dynamically updating object pools across turns. These two stages enable two novel object centric metrics tailored for multi turn evaluation and one global metric of visual quality: 1) EdiVal-IF, which measures instruction following by combining open vocabulary object detectors for symbolic checks with VLMs for semantic verification on detector guided crops; 2) EdiVal-CC, which evaluates content consistency by calculating semantic similarity of unchanged objects and background using the evolving object pools; and 3) EdiVal-VQ, which quantifies changes in overall visual quality with human preference models. Instantiating this pipeline, we build EdiVal Bench, a multi-turn editing benchmark covering 9 instruction types and 16 state-of-the-art editing models, spanning in-context, flow-matching, and diffusion paradigms. We demonstrate that EdiVal can be used to identify existing failure modes, thereby informing the development of the next generation of editing models.
format Preprint
id arxiv_https___arxiv_org_abs_2509_13399
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle EdiVal-Agent: An Object-Centric Framework for Automated, Fine-Grained Evaluation of Multi-Turn Editing
Chen, Tianyu
Zhang, Yasi
Zhang, Zhi
Yu, Peiyu
Wang, Shu
Wang, Zhendong
Lin, Kevin
Wang, Xiaofei
Yang, Zhengyuan
Li, Linjie
Lin, Chung-Ching
Xie, Jianwen
Leong, Oscar
Wang, Lijuan
Wu, Ying Nian
Zhou, Mingyuan
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Instruction-based image editing has advanced rapidly, yet reliable and interpretable evaluation remains a bottleneck. Current protocols either (i) depend on paired reference images, resulting in limited coverage and inheriting biases from prior generative models or (ii) rely solely on zero-shot vision language models (VLMs), whose prompt-based assessments of instruction following, content consistency, and visual quality are often imprecise. To address this, we introduce EdiVal, an automated and fine-grained evaluation framework grounded in an object-centric perspective, designed to assess not only standard single-turn but also multi-turn instruction-based editing with precision. Given an input image, EdiVal first decomposes it into semantically meaningful objects, then synthesizes diverse, context-aware editing instructions while dynamically updating object pools across turns. These two stages enable two novel object centric metrics tailored for multi turn evaluation and one global metric of visual quality: 1) EdiVal-IF, which measures instruction following by combining open vocabulary object detectors for symbolic checks with VLMs for semantic verification on detector guided crops; 2) EdiVal-CC, which evaluates content consistency by calculating semantic similarity of unchanged objects and background using the evolving object pools; and 3) EdiVal-VQ, which quantifies changes in overall visual quality with human preference models. Instantiating this pipeline, we build EdiVal Bench, a multi-turn editing benchmark covering 9 instruction types and 16 state-of-the-art editing models, spanning in-context, flow-matching, and diffusion paradigms. We demonstrate that EdiVal can be used to identify existing failure modes, thereby informing the development of the next generation of editing models.
title EdiVal-Agent: An Object-Centric Framework for Automated, Fine-Grained Evaluation of Multi-Turn Editing
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2509.13399