MICON-Bench: 多图像上下文图像生成的基准测试与增强
MICON-Bench: Benchmarking and Enhancing Multi-Image Context Image Generation in Unified Multimodal Models
AI总结：
MICON-Bench通过六个任务评估多图像上下文生成能力，并提出DAR机制提升生成质量与连贯性。
摘要：
近年来，统一多模态模型（UMMs）的进步使图像理解和生成能力有了显著提升。然而，尽管像Gemini-2.5-Flash-Image这样的模型展现出对多张相关图像进行推理的能力，现有基准测试很少解决多图像上下文生成的挑战，主要集中在文本到图像或单图像编辑任务上。在本工作中，我们介绍了MICON-Bench，一个涵盖六个任务的全面基准测试，评估跨图像组成、上下文推理和身份保持能力。我们进一步提出了一种基于MLLM的Evaluation-by-Checkpoint框架，用于自动验证语义和视觉一致性，其中多模态大语言模型（MLLM）充当验证者。此外，我们还提出了动态注意力再平衡（DAR），一种无需训练的即插即用机制，能够在推理过程中动态调整注意力以提高连贯性和减少幻觉。在各种最先进的开源模型上的广泛实验表明，MICON-Bench在暴露多图像推理挑战方面的严谨性以及DAR在提高生成质量和跨图像连贯性方面的有效性。Github: https://github.com/Angusliuuu/MICON-Bench。
英文摘要：
Recent advancements in Unified Multimodal Models (UMMs) have enabled remarkable image understanding and generation capabilities. However, while models like Gemini-2.5-Flash-Image show emerging abilities to reason over multiple related images, existing benchmarks rarely address the challenges of multi-image context generation, focusing mainly on text-to-image or single-image editing tasks. In this work, we introduce \textbf{MICON-Bench}, a comprehensive benchmark covering six tasks that evaluate cross-image composition, contextual reasoning, and identity preservation. We further propose an MLLM-driven Evaluation-by-Checkpoint framework for automatic verification of semantic and visual consistency, where multimodal large language model (MLLM) serves as a verifier. Additionally, we present \textbf{Dynamic Attention Rebalancing (DAR)}, a training-free, plug-and-play mechanism that dynamically adjusts attention during inference to enhance coherence and reduce hallucinations. Extensive experiments on various state-of-the-art open-source models demonstrate both the rigor of MICON-Bench in exposing multi-image reasoning challenges and the efficacy of DAR in improving generation quality and cross-image coherence. Github: https://github.com/Angusliuuu/MICON-Bench.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
