UFO：多模态图像生成中全条件对齐的评估链

Computer Science > Computer Vision and Pattern Recognition arXiv:2609.12397 (cs) [Submitted on 11 Sep 2026 (v1), last revised 17 Sep 2026 (this version, v2)] Title:UFO: Chain-of-Evaluation for Omni-Condition Alignment in Multi-Modal Image Generation Authors:Danning Zhang, Yijing Lin, Shuhan Zhuang, Mengqi Huang, Shaojin Wu, Shancheng Fang, Zhendong Mao View a PDF of the paper titled UFO: Chain-of-Evaluation for Omni-Condition Alignment in Multi-Modal Image Generation, by Danning Zhang and 6 other authors View PDF HTML (experimental) Abstract:Multi-modal image generation, particularly subject-driven customization, has garnered growing attention in recent years. Despite the rapid advancement of generative models, their evaluation remains largely lagging. Existing methods, whether embedding-based or Multi-modal Large Language Model (MLLM)-based, evaluate alignment with each modal condition in isolation, which contradicts the simultaneous condition alignment objective of multi-modal image generation, leading to poor consistency with human judgments. To address this challenge, we propose UFO, the first unified framework for omni-condition alignment simultaneous evaluation. Specifically, UFO introduces a novel Atomized Chain-of-Evaluation paradigm, i.e., it first decomposes omni-condition alignment into a sequential chain of fine-grained, disentangled Atomic Evaluation Units (AEUs), categorizes them into distinct modality-relevance classes, and then employs general or dedicated functional calls for accurate verification of different AEU types. Experimental results demonstrate that UFO achieves the highest correlation with human evaluation preferences, delivering an average improvement of 15.25%. Furthermore, we present UFO-Bench, a dedicated benchmark designed to holistically evaluate the performance of existing customization models under the diverse mutual interactions of textual and visual conditions. Comments: Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) Cite as: arXiv:2609.12397 [cs.CV] (or arXiv:2609.12397v2 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2609.12397 Focus to learn more arXiv-issued DOI via DataCite Journal reference: Proceedings of the Forty-Third International Conference on Machine Learning, 2026 Submission history From: Danning Zhang [view email] [v1] Fri, 11 Sep 2026 03:40:11 UTC (26,707 KB) [v2] Thu, 17 Sep 2026 13:13:59 UTC (26,711 KB) Full-text links: Access Paper: View a PDF of the paper titled UFO: Chain-of-Evaluation for Omni-Condition Alignment in Multi-Modal Image Generation, by Danning Zhang and 6 other authors View PDF HTML (experimental) TeX Source view license Current browse context: cs.CV new | recent | 2026-09 Change to browse by: cs cs.AI References & Citations NASA ADS Google Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv (What is alphaXiv?) Links to Code Toggle CatalyzeX Code Finder for Papers (What is CatalyzeX?) DagsHub Toggle DagsHub (What is DagsHub?) GotitPub Toggle Gotit.pub (What is GotitPub?) Huggingface Toggle Hugging Face (What is Huggingface?) ScienceCast Toggle ScienceCast (What is ScienceCast?) Demos Demos Replicate Toggle Replicate (What is Replicate?) Spaces Toggle Hugging Face Spaces (What is Spaces?) Spaces Toggle TXYZ.AI (What is TXYZ.AI?) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower (What are Influence Flowers?) Core recommender toggle CORE Recommender (What is CORE?) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs. Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)