GenRouter：面向智能体图像生成的统一工作流路由
GenRouter: Unified Workflow Routing for Agentic Image Generation
浏览论文内容
中文总结 AI 辅助
GenRouter是首个智能体图像生成的统一工作流路由框架，通过需求分析、经验匹配、帕累托过滤实现自适应路由，可大幅降低计算开销与延迟并提升视觉对齐效果。
中文摘要 AI 辅助
文本到图像（T2I）生成模型的快速发展已有效解决了原始像素合成的基础挑战，促使研究界将重点转向满足日益复杂的用户请求。尽管近期的智能体图像生成工作流通过外部知识检索、迭代推理等高级能力增强了静态推理，但它们大多以孤立的形式运行，采用固定的“一刀切”拓扑结构，不可避免地导致严重的计算不匹配问题——简单查询也被迫经过计算密集型的流水线。为弥合这一差距，本文提出GenRouter，首个面向智能体图像生成的统一工作流路由框架。我们首先构建GenCanvas，将不同的智能体流水线标准化为一组通用的基础原语和可执行模板。在该统一空间上运行时，GenRouter通过以下三种方式将异构提示自适应路由至最优工作流：（i）需求分析、（ii）经验匹配、（iii）帕累托过滤。在多个基准上开展的大量实验表明，与重量级静态流水线相比，GenRouter在实现更优视觉对齐的同时，将执行成本降低了95%以上，延迟降低了65%。此外，该系统可通过积累的经验持续自我进化，实现稳健的零样本泛化，进而提升性能并将计算开销减半。
英文摘要
The rapid evolution of text-to-image (T2I) generation models has effectively solved the foundational challenge of raw pixel synthesis, shifting the community's focus toward fulfilling increasingly intricate user requests. While recent agentic image generation workflows enhance static inference with advanced capabilities like external knowledge retrieval and iterative reasoning, they mostly operate in isolated silos with fixed ``one-size-fits-all" topologies. This inevitably leads to severe compute-mismatch, where simple queries are forced through computationally heavy pipelines. To bridge this gap, we present GenRouter, the first unified workflow routing framework for agentic image generation. We first formulate GenCanvas, standardizing diverse agentic pipelines into a universal set of foundational primitives and executable templates. Operating over this unified space, GenRouter adaptively routes heterogeneous prompts to their optimal workflows via (i) demand profiling, (ii) experience matching, and (iii) Pareto filtering. Extensive experiments across diverse benchmarks demonstrate that GenRouter achieves superior visual alignment while reducing execution costs by over 95% and latency by 65% compared to heavyweight static pipelines. Furthermore, the system continuously self-evolves via accumulated experience, enabling robust zero-shot generalization that boosts performance and halves computational overhead.
发表机构
- HKUST(GZ)（香港科技大学（广州））
- HKUST（香港科技大学）
- SUSTech（南方科技大学）
- ZODA
- CUHK（香港中文大学）
机构由 AI 辅助整理，请以论文原文为准。
