MemSlides：一個階層式記憶驅動的代理框架，用於具備多輪局部修訂的個人化簡報生成

摘要

個人化簡報生成不僅需要依賴當前的提示或模板來進行條件控制：代理必須在任務之間保持穩定的使用者偏好，在多輪修訂過程中保留新引入的偏好與限制，並可靠地執行局部編輯。我們提出 MemSlides，這是一個專為個人化簡報代理設計的分層記憶框架，將長期記憶與工作記憶分離，並進一步將長期記憶分為使用者設定檔記憶與工具記憶。使用者設定檔記憶儲存以意圖為條件的設定檔，用於第零輪的個人化設定；工作記憶則跨修訂輪次攜帶當前偏好與會話限制；工具記憶則儲存可重複使用的執行經驗，以達成可靠的局部編輯。MemSlides 將此記憶設計與有範圍限制的投影片局部修訂相結合，使得目標更新僅作用於最小的受影響區域，而非重複重新生成整個簡報。在控制實驗中，使用者設定檔記憶提升了在多角色、多意圖設定檔庫中的角色一致性判斷；工具記憶注入改善了在診斷配對設定中的閉環修改行為；而質性案例則展示了工作記憶在偏好延續方面的能力。綜合這些結果顯示，在簡報創作中實現有效的個人化，取決於將持久性的使用者設定檔、會話層級的工作記憶以及可重複使用的執行經驗，在生成與局部修訂過程中進行分離處理。

English

Personalized presentation generation requires more than conditioning on a current prompt or template: agents must preserve stable user preferences across tasks, retain newly introduced preferences and constraints during multi-turn revision, and carry out local edits reliably. We propose MemSlides, a hierarchical memory framework for personalized presentation agents that separates long-term memory from working memory and further divides long-term memory into user profile memory and tool memory. User profile memory stores intent-conditioned profiles for round-0 personalization, working memory carries active preferences and session constraints across revision rounds, and tool memory stores reusable execution experience for reliable localized editing. MemSlides pairs this memory design with scoped slide-local revision, so targeted updates act on the smallest affected region instead of repeatedly regenerating the full deck. In controlled experiments, user profile memory improves persona-alignment judgments on a multi-persona, multi-intent profile bank, tool-memory injection improves closed-loop modify behavior in diagnostic matched-pair settings, and qualitative cases illustrate working memory's ability to carryover preferences. Taken together, these results suggest that effective personalization in presentation authoring depends on separating persistent user profiles, session-level working memory, and reusable execution experience across generation and localized revision.