SlideGen: Collaborative Multimodal Agents for Scientific Slide Generation
Researchers have developed SlideGen, a system for generating presentation slides from scientific papers. Unlike previous approaches that focus on textual content selection, SlideGen uses a collaborative framework to assign specialized agents to tasks such as narrative planning, multimodal grounding, and layout composition. The system produces slide decks that are both faithful to the source paper and effective as presentations, with improved layout balance, content coverage,
Researchers have developed SlideGen, a system for generating presentation slides from scientific papers. Unlike previous approaches that focus on textual content selection, SlideGen uses a collaborative framework to assign specialized agents to tasks such as narrative planning, multimodal grounding, and layout composition. The system produces slide decks that are both faithful to the source paper and effective as presentations, with improved layout balance, content coverage, and text coherence.
---
Why it matters: Engineers working on AI-powered scientific communication tools will be interested in SlideGen's approach to multimodal design reasoning and agent collaboration, which can help improve the effectiveness of presentation slides generated from scientific papers.
Source: https://arxiv.org/abs/2512.04529
This article was originally published at: https://arxiv.org/abs/2512.04529