PaperVoyager : Building Interactive Web with Visual Language Models

TL;DR

PaperVoyager is a framework that converts research papers into interactive web systems, enabling dynamic exploration of mechanisms and interactions directly from PDFs.

Contribution

It introduces a novel end-to-end system that transforms research papers into executable interactive web applications, with a new benchmark and structured generation approach.

Findings

PaperVoyager significantly improves interactive system quality.

The benchmark of 19 papers provides a new evaluation standard.

End-to-end processing enables dynamic manipulation of research content.

Abstract

Recent advances in visual language models have enabled autonomous agents for complex reasoning, tool use, and document understanding. However, existing document agents mainly transform papers into static artifacts such as summaries, webpages, or slides, which are insufficient for technical papers involving dynamic mechanisms and state transitions. In this work, we propose a Paper-to-Interactive-System Agent that converts research papers into executable interactive web systems. Given a PDF paper, the agent performs end-to-end processing without human intervention, including paper understanding, system modeling, and interactive webpage synthesis, enabling users to manipulate inputs and observe dynamic behaviors. To evaluate this task, we introduce a benchmark of 19 research papers paired with expert-built interactive systems as ground truth. We further propose PaperVoyager, a structured generation framework that explicitly models mechanisms and interaction logic during synthesis. Experiments show that PaperVoyager significantly improves the quality of generated interactive systems, offering a new paradigm for interactive scientific paper understanding.