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

P2P: Automated Paper-to-Poster Generation and Fine-Grained Benchmark. Sun, T., Pan, E., Yang, Z., Sui, K., Shi, J., Cheng, X., Li, T., Huang, W., Zhang, G., Yang, J., & Li, Z. May, 2025. arXiv:2505.17104 [cs]

Paper doi abstract bibtex

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.

@misc{sun_p2p_2025,
	title = {{P2P}: {Automated} {Paper}-to-{Poster} {Generation} and {Fine}-{Grained} {Benchmark}},
	shorttitle = {{P2P}},
	url = {http://arxiv.org/abs/2505.17104},
	doi = {10.48550/arXiv.2505.17104},
	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.},
	urldate = {2026-02-05},
	publisher = {arXiv},
	author = {Sun, Tao and Pan, Enhao and Yang, Zhengkai and Sui, Kaixin and Shi, Jiajun and Cheng, Xianfu and Li, Tongliang and Huang, Wenhao and Zhang, Ge and Yang, Jian and Li, Zhoujun},
	month = may,
	year = {2025},
	note = {arXiv:2505.17104 [cs]},
	keywords = {Computer Science - Computation and Language, Computer Science - Multimedia},
}

Downloads: 0

{"_id":"h4rjpB4qs3CdaAscR","bibbaseid":"sun-pan-yang-sui-shi-cheng-li-huang-etal-p2pautomatedpapertopostergenerationandfinegrainedbenchmark-2025","author_short":["Sun, T.","Pan, E.","Yang, Z.","Sui, K.","Shi, J.","Cheng, X.","Li, T.","Huang, W.","Zhang, G.","Yang, J.","Li, Z."],"bibdata":{"bibtype":"misc","type":"misc","title":"P2P: Automated Paper-to-Poster Generation and Fine-Grained Benchmark","shorttitle":"P2P","url":"http://arxiv.org/abs/2505.17104","doi":"10.48550/arXiv.2505.17104","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.","urldate":"2026-02-05","publisher":"arXiv","author":[{"propositions":[],"lastnames":["Sun"],"firstnames":["Tao"],"suffixes":[]},{"propositions":[],"lastnames":["Pan"],"firstnames":["Enhao"],"suffixes":[]},{"propositions":[],"lastnames":["Yang"],"firstnames":["Zhengkai"],"suffixes":[]},{"propositions":[],"lastnames":["Sui"],"firstnames":["Kaixin"],"suffixes":[]},{"propositions":[],"lastnames":["Shi"],"firstnames":["Jiajun"],"suffixes":[]},{"propositions":[],"lastnames":["Cheng"],"firstnames":["Xianfu"],"suffixes":[]},{"propositions":[],"lastnames":["Li"],"firstnames":["Tongliang"],"suffixes":[]},{"propositions":[],"lastnames":["Huang"],"firstnames":["Wenhao"],"suffixes":[]},{"propositions":[],"lastnames":["Zhang"],"firstnames":["Ge"],"suffixes":[]},{"propositions":[],"lastnames":["Yang"],"firstnames":["Jian"],"suffixes":[]},{"propositions":[],"lastnames":["Li"],"firstnames":["Zhoujun"],"suffixes":[]}],"month":"May","year":"2025","note":"arXiv:2505.17104 [cs]","keywords":"Computer Science - Computation and Language, Computer Science - Multimedia","bibtex":"@misc{sun_p2p_2025,\n\ttitle = {{P2P}: {Automated} {Paper}-to-{Poster} {Generation} and {Fine}-{Grained} {Benchmark}},\n\tshorttitle = {{P2P}},\n\turl = {http://arxiv.org/abs/2505.17104},\n\tdoi = {10.48550/arXiv.2505.17104},\n\tabstract = {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.},\n\turldate = {2026-02-05},\n\tpublisher = {arXiv},\n\tauthor = {Sun, Tao and Pan, Enhao and Yang, Zhengkai and Sui, Kaixin and Shi, Jiajun and Cheng, Xianfu and Li, Tongliang and Huang, Wenhao and Zhang, Ge and Yang, Jian and Li, Zhoujun},\n\tmonth = may,\n\tyear = {2025},\n\tnote = {arXiv:2505.17104 [cs]},\n\tkeywords = {Computer Science - Computation and Language, Computer Science - Multimedia},\n}\n\n\n\n","author_short":["Sun, T.","Pan, E.","Yang, Z.","Sui, K.","Shi, J.","Cheng, X.","Li, T.","Huang, W.","Zhang, G.","Yang, J.","Li, Z."],"key":"sun_p2p_2025","id":"sun_p2p_2025","bibbaseid":"sun-pan-yang-sui-shi-cheng-li-huang-etal-p2pautomatedpapertopostergenerationandfinegrainedbenchmark-2025","role":"author","urls":{"Paper":"http://arxiv.org/abs/2505.17104"},"keyword":["Computer Science - Computation and Language","Computer Science - Multimedia"],"metadata":{"authorlinks":{}}},"bibtype":"misc","biburl":"https://bibbase.org/zotero-group/schulzkx/5158478","dataSources":["JFDnASMkoQCjjGL8E"],"keywords":["computer science - computation and language","computer science - multimedia"],"search_terms":["p2p","automated","paper","poster","generation","fine","grained","benchmark","sun","pan","yang","sui","shi","cheng","li","huang","zhang","yang","li"],"title":"P2P: Automated Paper-to-Poster Generation and Fine-Grained Benchmark","year":2025}