# 逆向规划实现个性化：通过结构去噪学习潜在设计意图的智能体幻灯片生成

- 来源：HuggingFace Daily Papers（社区热门论文）
- 发布时间：2026-07-01 08:00
- AIHOT 分数：51
- AIHOT 链接：https://aihot.virxact.com/items/cmr38srx801mxsly0m85k240o
- 原文链接：https://arxiv.org/abs/2607.00407

## AI 摘要

幻灯片个性化需要同时定制主题与布局，现有AI智能体方法依赖预设模板或用户详细指令，难以捕捉细粒度潜在设计意图。SPIRE将页面级幻灯片个性化（PSP）重新定义为逆向规划问题，在不假设具体执行工具（如PowerPoint、Beamer）的前提下学习设计意图。通过故意破坏干净幻灯片的视觉结构，SPIRE创建可验证的去噪任务，两个智能体通过强化学习协作优化可执行设计。理论证明结构去噪是PSP的一致代理，且多智能体公式严格降低策略梯度方差。实验表明SPIRE在幻灯片个性化生成上表现优越。

## 正文

Slide design requires personalizing both deck themes and page layouts. Yet, current AI agent-based methods struggle with fine-grained, page-level design. Solely relying on prespecified templates or user verbose instructions, they fail to capture latent design intents, leaving Page-level Slide Personalization (PSP) unresolved. To close this gap, this work formulates PSP as an inverse planning problem. We propose to learn a design intent without assuming any knowledge of the specific executing tools (e.g., PowerPoint, Beamer) being used. However, relinquishing control over these tools makes the problem intractable to optimize end-to-end. To overcome this, we propose SPIRE, a principled framework to solve PSP approximately. By intentionally corrupting the visual structures of clean slides, SPIRE creates a verifiable task to denoise the corruption, whereby two agents learn to collaboratively refine executable designs via reinforcement learning (RL). We present a proof that structural denoising is a consistent surrogate for PSP, and that the multi-agent formulation strictly reduces policy gradient variance in RL. Extensive experiments demonstrate the superiority of SPIRE.