{"ID":2889827,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.21391","arxiv_id":"2507.21391","title":"Multimodal LLMs as Customized Reward Models for Text-to-Image Generation","abstract":"We introduce LLaVA-Reward, an efficient reward model designed to automatically evaluate text-to-image (T2I) generations across multiple perspectives, leveraging pretrained multimodal large language models (MLLMs). Existing MLLM-based approaches require instruction-following data for supervised fine-tuning and evaluate generation quality on analyzing text response, which is time-consuming and difficult to train. To address this problem, we propose LLaVA-Reward, which directly utilizes the hidden states of MLLMs given text-image pairs. To enhance the bidirectional interaction between visual and textual representations in decoder-only MLLMs, we further propose adding a Skip-connection Cross Attention (SkipCA) module. This design enhances text-image correlation reasoning by connecting early-layer visual features with later-layer hidden representations. In addition, LLaVA-Reward supports different types of preference data for efficient fine-tuning, including paired preference data and unpaired data. We train LLaVA-Reward on four evaluation perspectives: text-image alignment, fidelity/artifact, safety, and overall ranking. Empirical results demonstrate that LLaVA-Reward outperforms conventional and MLLM-based methods in generating human-aligned scores for automatic evaluations and inference-time scaling in text-to-image generations.","short_abstract":"We introduce LLaVA-Reward, an efficient reward model designed to automatically evaluate text-to-image (T2I) generations across multiple perspectives, leveraging pretrained multimodal large language models (MLLMs). Existing MLLM-based approaches require instruction-following data for supervised fine-tuning and evaluate...","url_abs":"https://arxiv.org/abs/2507.21391","url_pdf":"https://arxiv.org/pdf/2507.21391v2","authors":"[\"Shijie Zhou\",\"Ruiyi Zhang\",\"Huaisheng Zhu\",\"Branislav Kveton\",\"Yufan Zhou\",\"Jiuxiang Gu\",\"Jian Chen\",\"Changyou Chen\"]","published":"2025-07-28T23:52:53Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.AI\",\"cs.CL\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}