发表机构
Tencent; Peking University（腾讯; 北京大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究针对统一多模态大语言模型在可控图像生成中难以遵循复杂空间指令的问题，提出ATLAS框架，采用“思考、规划、绘制”范式及布局共享表示，经强化学习提升性能，在图像生成等任务中效果显著，还支持相关编辑与多模态基础，并引入评估基准。
AI 中文摘要
统一多模态大语言模型（MLLMs）为统一视觉理解和生成提供了有前景的范式，但在可控图像生成中难以遵循复杂空间指令和逻辑约束。为解决此差距，我们提出ATLAS，一个为MLLMs配备类人“思考、规划、绘制”范式的统一框架。采用布局作为连接三个阶段的共享表示，使模型能推理空间需求、规划明确物体排列并渲染最终图像。通过基于强化学习的布局对齐进一步提高规划到图像的保真度。我们在7B和80B规模实例化ATLAS，在图像生成基准测试中取得MLLMs的最优性能，并比现有基于布局的统一MLLMs平均提高65.31%。在空间相关任务上，ATLAS比基础模型平均提升23.06%。通过相同布局接口，ATLAS还支持指令引导编辑和多模态基础。我们还引入ATLAS-Reasoning，一个用于评估复杂空间指令下生成的基准。
英文摘要
Unified Multimodal Large Language Models (MLLMs) offer a promising paradigm for unifying visual understanding and generation, yet they still struggle to follow complex spatial instructions and logical constraints in controllable image generation. To address this gap, we present ATLAS, a unified framework that equips MLLMs with a human-like "Think, Plan, and Paint" paradigm. We adopt layout as the shared representation that connects the three stages, enabling the model to reason about spatial requirements, plan explicit object arrangements, and render the final image. We further improve plan-to-image fidelity with reinforcement-learning-based layout alignment. We instantiate ATLAS at 7B and 80B scales, achieving state-of-the-art performance among MLLMs on image generation benchmarks and an average 65.31% improvement over existing layout-based unified MLLMs. On spatially related tasks, ATLAS obtains an average 23.06% gain over the base models. Through the same layout interface, ATLAS also supports instruction-guided editing and multimodal grounding. We further introduce ATLAS-Reasoning, a benchmark for evaluating generation under complex spatial instructions.
Comments22 pages, 12 figures, 9 tables
