[论文] AesCode: Aesthetic Code Generation with Decoupled Cross-Modal Rewards

论文概要

研究领域: ML 作者: Xin Gao, Yifan Yang, Xin Li, Pan Wang, Yan Li, Muzhao Tian, Mingxi Cheng, Bei Liu, Kai Qiu, Qi Dai, Dong Chen, Yue Dong, Leandros Tassiulas, Rex Ying, Ji Li, Chong Luo 发布时间: 2026-10-13 arXiv: 2610.18860

中文摘要

以代码形式生成幻灯片、海报和仪表盘等信息密集型视觉产物很有吸引力，因为输出结构化、可编辑且可验证。然而代码模型无法看到其布局和颜色选择如何在画布上呈现，往往限制了空间规划和美学质量。图像生成器擅长视觉设计但可能错误渲染请求的内容，使其输出不可靠。因此我们提出 AesCode：生成以内容提示为根基的代码，同时使用从同一提示生成的图像作为美学引导。AesCode 将目标表示为描述整个画布的设计图，使每个属性可归因。从此图推导出互补的跨模态奖励：确定性验证器为可从代码和渲染解析的属性打分，VLM 评判器使用与图元素和关系绑定的样本特定视觉图评分标准（VGR）评估不可解析的属性。我们通过带逐通道归一化的强化学习优化这些解耦奖励，联合改进空间规划、美学质量和可验证正确性。为评估此场景，我们构建了分别评分可验证正确性和视觉质量的基准，视觉分数与人类判断高度一致。我们发现参考图像显著改善了现有模型生成代码的视觉质量，消融实验表明 AesCode 同时从参考图像作为模型输入和视觉奖励来源中受益。相对于具有相同参考的同等规模 Qwen3-VL 基线，AesCode-8B 和 AesCode-32B 分别将视觉质量提升 31.1 和 22.4 个百分点，达到与 GPT-5.5 和 Claude Opus 4.8 等闭源前沿模型相当的分数。

原文摘要

Generating information-rich visual artifacts such as slides, posters, and dashboards as code is appealing because the outputs are structured, editable, and verifiable. Yet code models cannot see how their choices of layout and color come together on the canvas, often limiting spatial planning and aesthetic quality. Image generators excel at visual design but can misrender requested content, making their outputs unreliable targets for direct reproduction. We therefore introduce AesCode, which generates code grounded in the content prompt while using an image generated from the same prompt as an aesthetic guidance. To do so, AesCode represents the target as a design graph that describes the whole canvas and makes each property attributable. From this graph, we derive complementary cross-moda...

#论文 #arXiv #ML #小凯