Search papers, labs, and topics across Lattice.
This paper introduces the Multi-dimensional Evaluation-Verification Reward (EVR) framework to enhance multi-reference image editing by addressing the challenges of visual consistency and harmony across multiple references. By employing a reinforcement learning approach that utilizes a multimodal large language model (MLLM) to generate and verify candidate hypotheses against visual evidence, the method produces reliable reward signals for fine-tuning image editors. Experimental results demonstrate that EVR significantly improves the performance of existing models, achieving consistency and harmony levels that match or exceed those of state-of-the-art systems like NanoBanana.
Multi-dimensional Evaluation-Verification Reward transforms multi-reference image editing by providing a structured approach to evaluate and enhance visual consistency, yielding superior results over existing models.
While recent image editing models have made rapid progress, multi-reference editing remains challenging, particularly in maintaining visual consistency across references and ensuring overall visual harmony. Reinforcement learning has proven highly effective for text-to-image generation and single-image editing, but its extension to multi-reference editing is hindered by the absence of suitable reward models that capture multi-image relational constraints. Moreover, naively using multimodal large language models(MLLMs) as zero-shot evaluators faces a key tension between hallucination-prone long-form reasoning and the limited deductive power of short-form judgments. We address these issues with a Multi-dimensional Evaluation-Verification Reward(EVR). EVR decomposes evaluation into distinct visual criteria; for each criterion, an MLLM Evaluator generates multiple candidate hypotheses, and a Verifier grounds each claim in concrete visual evidence to accept or reject it, producing reliable and fine-grained reward signals. Together with a scalable data pipeline, our method enables RL fine-tuning of off-the-shelf editors without architectural changes. Extensive experiments show substantial gains over the base Qwen-Image-Edit, improving consistency and harmony to match or surpass NanoBanana.
