AutoDesign：面向长视距智能体设计的元工具优化
AutoDesign: Meta-Harness Optimization for Long-Horizon Agentic Design
- Meituan（美团）
- MBZUAI（Mohamed bin Zayed University of Artificial Intelligence）
- Huazhong University of Science and Technology（华中科技大学）
- Peking University（北京大学）
- Tsinghua University（清华大学）
- The Chinese University of Hong Kong（香港中文大学）
- Shanghai Jiao Tong University（上海交通大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
AutoDesign是符合人类设计先验的元工具优化框架，以论文转海报生成任务为实例，在PosterBench上性能优于Claude Design，集成其学习的DesignHarness可提升代码智能体性能，且获人类最高偏好。
AI中文摘要：
将多模态源转化为浓缩且结构化的媒体输出，可从根本上被概念化为以模型-工具系统为核心的长视距智能体过程。理想的工具系统应符合人类设计先验，并通过经验探索积累可复用的经验以驱动递归自我改进，而现有范式仍为静态，缺乏此能力。本文提出AutoDesign框架，其符合人类设计先验，其中元工具优化器指导代码智能体基于rollout反馈递归改进工具。为实例化和评估该框架，我们聚焦学术论文到海报的生成任务，引入PosterBench，包含覆盖五个学科的100篇论文主赛道，以及用于受控评估的共享10篇论文子集PosterBench-mini。在PosterBench主赛道上，AutoDesign取得78.32的最高分，超过闭源商业系统Claude Design 7.45分。在七个受控代码智能体-模型配置中，集成学习得到的DesignHarness始终提升性能，将平均PosterBench分数从54.99提高至67.39（+12.4%）。在完全自主的长视距循环中，它在40分钟内执行253次工具调用和11次编辑轮次，成本低于3美元，在人类评估中达到会议海报的平均质量；系统盲测人类研究进一步表明，AutoDesign在被评估系统中获得最高的人类偏好。
英文摘要：
Transforming multimodal sources into condensed and structured media outputs can be fundamentally conceptualized as a long-horizon agentic process centered on a model-harness system. While an ideal harness system should align with human design priors and accumulate reusable experience through empirical exploration to drive recursive self-improvement, existing paradigms remain static and fall short of this capability. In this paper, we present AutoDesign, a framework that aligns with human design priors, where a meta-harness optimizer guides a code agent to recursively improve harness based on rollout feedback. To instantiate and evaluate this framework, we focus on the academic paper-to-poster generation task and introduce PosterBench, comprising a 100-paper Main Track spanning five disciplines and PosterBench-mini, a shared 10-paper subset for controlled evaluation. On the PosterBench Main Track, AutoDesign achieves the highest score of 78.32, surpassing the closed-source commercial system Claude Design by 7.45 points. Across seven controlled code-agent-model configurations, integrating the learned DesignHarness consistently improves performance, increasing the average PosterBench Score from 54.99 to 67.39 (+12.4%). In a fully autonomous long-horizon loop, it executes 253 tool calls and 11 editing turns within 40 minutes for under $3, reaching average conference-poster quality in human evaluation. A system-blind human study further demonstrates that AutoDesign achieves the highest human preference among evaluated systems.
