Paper2Web: Let's Make Your Paper Alive!

TL;DR

Paper2Web addresses the challenge of converting static academic PDFs into interactive, multimedia-rich homepages that preserve core content. It introduces the Paper2Web task and dataset, linking papers to verified project homepages (10,716 pairs), and establishes a comprehensive evaluation suite with Connectivity, Completeness, a holistic MLLM-as-a-Judge, and PaperQuiz. PWAgent, a MCP-based multi-agent pipeline, decomposes papers into structured assets, ingests them into an MCP Resource Repository, and iteratively refines layouts and content to produce high-quality pages at low cost. Across extensive experiments, PWAgent outperforms end-to-end generation and arXiv/alphaXiv baselines on multiple metrics, approaching ground-truth aesthetics and informativeness while delivering strong interactivity and balanced multimedia presentation. Overall, this work advances scalable, automated dissemination of scholarly work by enabling rich, navigable, and knowledge-transfer-focused academic webpages.

Abstract

Academic project websites can more effectively disseminate research when they clearly present core content and enable intuitive navigation and interaction. However, current approaches such as direct Large Language Model (LLM) generation, templates, or direct HTML conversion struggle to produce layout-aware, interactive sites, and a comprehensive evaluation suite for this task has been lacking. In this paper, we introduce Paper2Web, a benchmark dataset and multi-dimensional evaluation framework for assessing academic webpage generation. It incorporates rule-based metrics like Connectivity, Completeness and human-verified LLM-as-a-Judge (covering interactivity, aesthetics, and informativeness), and PaperQuiz, which measures paper-level knowledge retention. We further present PWAgent, an autonomous pipeline that converts scientific papers into interactive and multimedia-rich academic homepages. The agent iteratively refines both content and layout through MCP tools that enhance emphasis, balance, and presentation quality. Our experiments show that PWAgent consistently outperforms end-to-end baselines like template-based webpages and arXiv/alphaXiv versions by a large margin while maintaining low cost, achieving the Pareto-front in academic webpage generation.

Figures (15)

• Figure 1: Our work, Paper2Web, constitutes an important piece of the puzzle for the presentation and dissemination of academic papers. We build a unified platform to streamline all academic presentation at https://github.com/YuhangChen1/Paper2All.
• Figure 2: Problems in current scholar web page generation, including distorted layout and limited interactivity.
• Figure 3: Pareto-front comparison of each website generation methods. Our PWAgent achieve the highest quality with moderate and affordable cost.
• Figure 4: To transform static papers into exploratory web pages, we collect the first paper-webpage dataset by crawling across multiple top-tier conferences and filtering by online search and human annotators.
• Figure 5: Paper2Web dataset statistics.