Open Multimodal Retrieval-Augmented Factual Image Generation
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866918171476230144 |
|---|---|
| author | Tian, Yang Liu, Fan Zhang, Jingyuan Bi, Wei Hu, Yupeng Nie, Liqiang |
| author_facet | Tian, Yang Liu, Fan Zhang, Jingyuan Bi, Wei Hu, Yupeng Nie, Liqiang |
| contents | Large Multimodal Models (LMMs) have achieved remarkable progress in generating photorealistic and prompt-aligned images, but they often produce outputs that contradict verifiable knowledge, especially when prompts involve fine-grained attributes or time-sensitive events. Conventional retrieval-augmented approaches attempt to address this issue by introducing external information, yet they are fundamentally incapable of grounding generation in accurate and evolving knowledge due to their reliance on static sources and shallow evidence integration. To bridge this gap, we introduce ORIG, an agentic open multimodal retrieval-augmented framework for Factual Image Generation (FIG), a new task that requires both visual realism and factual grounding. ORIG iteratively retrieves and filters multimodal evidence from the web and incrementally integrates the refined knowledge into enriched prompts to guide generation. To support systematic evaluation, we build FIG-Eval, a benchmark spanning ten categories across perceptual, compositional, and temporal dimensions. Experiments demonstrate that ORIG substantially improves factual consistency and overall image quality over strong baselines, highlighting the potential of open multimodal retrieval for factual image generation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_22521 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Open Multimodal Retrieval-Augmented Factual Image Generation Tian, Yang Liu, Fan Zhang, Jingyuan Bi, Wei Hu, Yupeng Nie, Liqiang Computer Vision and Pattern Recognition Artificial Intelligence Information Retrieval Machine Learning Large Multimodal Models (LMMs) have achieved remarkable progress in generating photorealistic and prompt-aligned images, but they often produce outputs that contradict verifiable knowledge, especially when prompts involve fine-grained attributes or time-sensitive events. Conventional retrieval-augmented approaches attempt to address this issue by introducing external information, yet they are fundamentally incapable of grounding generation in accurate and evolving knowledge due to their reliance on static sources and shallow evidence integration. To bridge this gap, we introduce ORIG, an agentic open multimodal retrieval-augmented framework for Factual Image Generation (FIG), a new task that requires both visual realism and factual grounding. ORIG iteratively retrieves and filters multimodal evidence from the web and incrementally integrates the refined knowledge into enriched prompts to guide generation. To support systematic evaluation, we build FIG-Eval, a benchmark spanning ten categories across perceptual, compositional, and temporal dimensions. Experiments demonstrate that ORIG substantially improves factual consistency and overall image quality over strong baselines, highlighting the potential of open multimodal retrieval for factual image generation. |
| title | Open Multimodal Retrieval-Augmented Factual Image Generation |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Information Retrieval Machine Learning |
| url | https://arxiv.org/abs/2510.22521 |