SlideTailor: Personalized Presentation Slide Generation for Scientific Papers

TL;DR

The paper tackles personalized automatic slide generation for scientific papers by formalizing user preferences through two inputs: a paper-slides example pair and a slide template. It introduces an agentic framework that distills implicit content and aesthetic preferences, plans slide content with a narrated outline, and realizes slides by editing templates, enabling downstream video presentations. A dedicated benchmark and interpretable metrics are built to evaluate alignment with user intent and slide quality. Experimental results show that the proposed approach outperforms baselines in both preference alignment and overall presentation quality, supporting broader applications in adaptive, multimodal document-to-slide generation.

Abstract

Automatic presentation slide generation can greatly streamline content creation. However, since preferences of each user may vary, existing under-specified formulations often lead to suboptimal results that fail to align with individual user needs. We introduce a novel task that conditions paper-to-slides generation on user-specified preferences. We propose a human behavior-inspired agentic framework, SlideTailor, that progressively generates editable slides in a user-aligned manner. Instead of requiring users to write their preferences in detailed textual form, our system only asks for a paper-slides example pair and a visual template - natural and easy-to-provide artifacts that implicitly encode rich user preferences across content and visual style. Despite the implicit and unlabeled nature of these inputs, our framework effectively distills and generalizes the preferences to guide customized slide generation. We also introduce a novel chain-of-speech mechanism to align slide content with planned oral narration. Such a design significantly enhances the quality of generated slides and enables downstream applications like video presentations. To support this new task, we construct a benchmark dataset that captures diverse user preferences, with carefully designed interpretable metrics for robust evaluation. Extensive experiments demonstrate the effectiveness of our framework.

Figures (21)

• Figure 1: Preference-guided paper-to-slides generation. Based on user preferences inferred from a paper-slides sample pair and a visual template, the system produces personalized slides accompanied by a speech script, supporting downstream applications such as video presentations.
• Figure 2: The conceptual pipeline of our proposed preference-guided paper-to-sildes generation framework.
• Figure 3: An example of distilled content preference.
• Figure 4: An example of distilled aesthetic preference.
• Figure 5: Example output from paper reorganizer (left), slide outline designer (middle), and template selector (right).