P2P: Automated Paper-to-Poster Generation and Fine-Grained Benchmark

2025-05-21

AI Summary

Manually designing academic posters is time-consuming, while existing automated approaches struggle to simultaneously preserve scientific fidelity and achieve coherent multimodal (text-image) representation, compounded by the absence of standardized evaluation benchmarks. This paper proposes the first LLM-driven multi-agent framework—comprising vision parsing, content generation, and HTML assembly agents, augmented by a verification module—for end-to-end conversion of research papers into high-quality HTML posters. We introduce P2PInstruct, the first large-scale instruction dataset (32K samples) for poster generation, and P2PEval, the first fine-grained bimodal evaluation benchmark comprising 121 paper-poster pairs. Our methodology integrates LLM-as-a-Judge automated assessment, human-annotated checklists, and iterative verification. Experiments demonstrate that our generated posters exhibit high semantic accuracy, structural compliance, and visual appeal, significantly improving both generation quality and evaluation reliability.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Multiagent Systems: Adversarial Agents
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Academic posters are vital for scholarly communication, yet their manual creation is time-consuming. However, automated academic poster generation faces significant challenges in preserving intricate scientific details and achieving effective visual-textual integration. Existing approaches often struggle with semantic richness and structural nuances, and lack standardized benchmarks for evaluating generated academic posters comprehensively. To address these limitations, we introduce P2P, the first flexible, LLM-based multi-agent framework that generates high-quality, HTML-rendered academic posters directly from research papers, demonstrating strong potential for practical applications. P2P employs three specialized agents-for visual element processing, content generation, and final poster assembly-each integrated with dedicated checker modules to enable iterative refinement and ensure output quality. To foster advancements and rigorous evaluation in this domain, we construct and release P2PInstruct, the first large-scale instruction dataset comprising over 30,000 high-quality examples tailored for the academic paper-to-poster generation task. Furthermore, we establish P2PEval, a comprehensive benchmark featuring 121 paper-poster pairs and a dual evaluation methodology (Universal and Fine-Grained) that leverages LLM-as-a-Judge and detailed, human-annotated checklists. Our contributions aim to streamline research dissemination and provide the community with robust tools for developing and evaluating next-generation poster generation systems.

Problem

• Automating academic poster creation from research papers
• Preserving scientific details in visual-textual integration
• Lacking standardized benchmarks for poster evaluation

Innovation

• LLM-based multi-agent framework for poster generation
• Specialized agents for visual and content processing
• Large-scale dataset and dual evaluation benchmark