@inproceedings{wang-etal-2025-mirror,
title = "Mirror in the Model: Ad Banner Image Generation via Reflective Multi-{LLM} and Multi-modal Agents",
author = "Wang, Zhao and
Chen, Bowen and
Shimose, Yotaro and
Moriyama, Sota and
Wang, Heng and
Takamatsu, Shingo",
editor = "Potdar, Saloni and
Rojas-Barahona, Lina and
Montella, Sebastien",
booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: Industry Track",
month = nov,
year = "2025",
address = "Suzhou (China)",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.emnlp-industry.17/",
pages = "246--266",
ISBN = "979-8-89176-333-3",
abstract = "Recent generative models such as GPT{-}4o have shown strong capabilities in producing high-quality images with accurate text rendering. However, commercial design tasks like advertising banners demand more than visual fidelity{---}they require structured layouts, precise typography, consistent branding and etc. In this paper, we introduce **MIMO (Mirror In{-}the{-}Model)**, an agentic refinement framework for automatic ad banner generation. MIMO combines a hierarchical multimodal agent system (MIMO{-}Core) with a coordination loop (MIMO{-}Loop) that explores multiple stylistic directions and iteratively improves design quality. Requiring only a simple natural language based prompt and logo image as input, MIMO automatically detects and corrects multiple types of errors during generation. Experiments show that MIMO significantly outperforms existing diffusion and LLM-based baselines in real-world banner design scenarios."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2025-mirror">
<titleInfo>
<title>Mirror in the Model: Ad Banner Image Generation via Reflective Multi-LLM and Multi-modal Agents</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zhao</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bowen</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yotaro</namePart>
<namePart type="family">Shimose</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sota</namePart>
<namePart type="family">Moriyama</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Heng</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shingo</namePart>
<namePart type="family">Takamatsu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: Industry Track</title>
</titleInfo>
<name type="personal">
<namePart type="given">Saloni</namePart>
<namePart type="family">Potdar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lina</namePart>
<namePart type="family">Rojas-Barahona</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sebastien</namePart>
<namePart type="family">Montella</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Suzhou (China)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-333-3</identifier>
</relatedItem>
<abstract>Recent generative models such as GPT-4o have shown strong capabilities in producing high-quality images with accurate text rendering. However, commercial design tasks like advertising banners demand more than visual fidelity—they require structured layouts, precise typography, consistent branding and etc. In this paper, we introduce **MIMO (Mirror In-the-Model)**, an agentic refinement framework for automatic ad banner generation. MIMO combines a hierarchical multimodal agent system (MIMO-Core) with a coordination loop (MIMO-Loop) that explores multiple stylistic directions and iteratively improves design quality. Requiring only a simple natural language based prompt and logo image as input, MIMO automatically detects and corrects multiple types of errors during generation. Experiments show that MIMO significantly outperforms existing diffusion and LLM-based baselines in real-world banner design scenarios.</abstract>
<identifier type="citekey">wang-etal-2025-mirror</identifier>
<location>
<url>https://aclanthology.org/2025.emnlp-industry.17/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>246</start>
<end>266</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Mirror in the Model: Ad Banner Image Generation via Reflective Multi-LLM and Multi-modal Agents
%A Wang, Zhao
%A Chen, Bowen
%A Shimose, Yotaro
%A Moriyama, Sota
%A Wang, Heng
%A Takamatsu, Shingo
%Y Potdar, Saloni
%Y Rojas-Barahona, Lina
%Y Montella, Sebastien
%S Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: Industry Track
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou (China)
%@ 979-8-89176-333-3
%F wang-etal-2025-mirror
%X Recent generative models such as GPT-4o have shown strong capabilities in producing high-quality images with accurate text rendering. However, commercial design tasks like advertising banners demand more than visual fidelity—they require structured layouts, precise typography, consistent branding and etc. In this paper, we introduce **MIMO (Mirror In-the-Model)**, an agentic refinement framework for automatic ad banner generation. MIMO combines a hierarchical multimodal agent system (MIMO-Core) with a coordination loop (MIMO-Loop) that explores multiple stylistic directions and iteratively improves design quality. Requiring only a simple natural language based prompt and logo image as input, MIMO automatically detects and corrects multiple types of errors during generation. Experiments show that MIMO significantly outperforms existing diffusion and LLM-based baselines in real-world banner design scenarios.
%U https://aclanthology.org/2025.emnlp-industry.17/
%P 246-266
Markdown (Informal)
[Mirror in the Model: Ad Banner Image Generation via Reflective Multi-LLM and Multi-modal Agents](https://aclanthology.org/2025.emnlp-industry.17/) (Wang et al., EMNLP 2025)
ACL