{"ID":2886579,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.04732","arxiv_id":"2508.04732","title":"LumiGen: An LVLM-Enhanced Iterative Framework for Fine-Grained Text-to-Image Generation","abstract":"Text-to-Image (T2I) generation has made significant advancements with diffusion models, yet challenges persist in handling complex instructions, ensuring fine-grained content control, and maintaining deep semantic consistency. Existing T2I models often struggle with tasks like accurate text rendering, precise pose generation, or intricate compositional coherence. Concurrently, Vision-Language Models (LVLMs) have demonstrated powerful capabilities in cross-modal understanding and instruction following. We propose LumiGen, a novel LVLM-enhanced iterative framework designed to elevate T2I model performance, particularly in areas requiring fine-grained control, through a closed-loop, LVLM-driven feedback mechanism. LumiGen comprises an Intelligent Prompt Parsing \u0026 Augmentation (IPPA) module for proactive prompt enhancement and an Iterative Visual Feedback \u0026 Refinement (IVFR) module, which acts as a \"visual critic\" to iteratively correct and optimize generated images. Evaluated on the challenging LongBench-T2I Benchmark, LumiGen achieves a superior average score of 3.08, outperforming state-of-the-art baselines. Notably, our framework demonstrates significant improvements in critical dimensions such as text rendering and pose expression, validating the effectiveness of LVLM integration for more controllable and higher-quality image generation.","short_abstract":"Text-to-Image (T2I) generation has made significant advancements with diffusion models, yet challenges persist in handling complex instructions, ensuring fine-grained content control, and maintaining deep semantic consistency. Existing T2I models often struggle with tasks like accurate text rendering, precise pose gene...","url_abs":"https://arxiv.org/abs/2508.04732","url_pdf":"https://arxiv.org/pdf/2508.04732v1","authors":"[\"Xiaoqi Dong\",\"Xiangyu Zhou\",\"Nicholas Evans\",\"Yujia Lin\"]","published":"2025-08-05T20:53:43Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.GR\"]","methods":"[\"Diffusion Model\",\"Language Model\"]","has_code":false}