Paper2Poster: Towards Multimodal Poster Automation from Scientific Papers
Wei Pang, Kevin Qinghong Lin, Xiangru Jian, Xi He, Philip Torr
2025-05-28
Summary
This paper is about creating a system that can automatically turn scientific papers into posters by using different types of information, like text and images, and then checking how good the posters are.
What's the problem?
The problem is that making posters from scientific papers takes a lot of time and effort, and it's hard to make sure the posters look good, make sense, and accurately represent the original content.
What's the solution?
The researchers built a new way to judge how well posters are made by looking at things like how they look, how well the information fits together, and if the content is correct. They also created a multi-step system, using several AI agents working together, that can make posters better and faster than older methods, while using less computer power.
Why it matters?
This matters because it makes it easier and quicker for scientists to share their work at conferences and events, helping more people understand and learn from new research without spending so much time designing posters by hand.
Abstract
A benchmark and metric suite for poster generation evaluates visual quality, coherence, and content accuracy, leading to a multi-agent pipeline that outperforms existing models with reduced computational cost.