Previous works have explored various customized generation tasks given a reference image, but they still face limitations in generating consistent fine-grained details. In this paper, our aim is to solve the inconsistency problem of generated images by applying a reference-guided post-editing approach and present our ImageCritic. We first construct a dataset of reference-degraded-target triplets obtained via VLM-based selection and explicit degradation, which effectively simulates the common inaccuracies or inconsistencies observed in existing generation models. Furthermore, building on a thorough examination of the model's attention mechanisms and intrinsic representations, we accordingly devise an attention alignment loss and a detail encoder to precisely rectify inconsistencies. ImageCritic can be integrated into an agent framework to automatically detect inconsistencies and correct them with multi-round and local editing in complex scenarios. Extensive experiments demonstrate that ImageCritic can effectively resolve detail-related issues in various customized generation scenarios, providing significant improvements over existing methods.
@article{arxiv.2511.20614,
title = {The Consistency Critic: Correcting Inconsistencies in Generated Images via Reference-Guided Attentive Alignment},
author = {Ziheng Ouyang and Yiren Song and Yaoli Liu and Shihao Zhu and Qibin Hou and Ming-Ming Cheng and Mike Zheng Shou},
journal= {arXiv preprint arXiv:2511.20614},
year = {2025}
}