AI 自己改自己的设计流程，论文海报做到 78 分
现在的 AI 做「把论文变成海报」这种长任务，流程是写死的：模型按固定套路跑，不会越做越熟。AutoDesign 让 AI 自己改自己的流程——它先按人类设计习惯搭一个初始框架，然后跑一遍、看结果、再让一个「元优化器」去改框架本身，如此循环。在 100 篇论文的测试里，它拿到 78.32 分，比闭源的商业系统 Claude Design 高 7.45 分；而且把学到的流程套到其他模型上，平均分从 54.99 涨到 67.39。最狠的是全自动跑完 253 次工具调用、11 轮修改，40 分钟、成本不到 3 美元，人类盲评里它也是首选。这不是你明天能用的功能，但它指向一个趋势：AI 不再只是执行任务，而是开始优化自己执行任务的方式——这比单次任务变强更值得注意。
📄 原文摘要(英文)
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.