The three of us -- Sebastian, Nolan, and Ian -- have really been getting into undergraduate research over the last year in various areas of machine learning. Despite being in different labs and reading different papers, we all noted similar difficulties when parsing through advanced topics in such a rapidly evolving field. One thing we can all agree on is that the experience of joining a new research lab, especially as an undergraduate with little experience, can feel intimidating, isolating, and occasionally unwelcoming. Students are often thrust into an unfamiliar role where for the first time in their educational career, there is little support and high expectations. We wondered if there was an easier way for students like us to begin making meaningful contributions to our labs and the computing community worldwide.
We wanted to create PaperBulb as a way to provide a pathway for new undergraduate researchers to bridge the gap between what they learn in their classes and what they need to know to be productive in their lab. Instead of stumbling through arbitrarily chosen papers and struggling to understand jargon, we wanted to create a tool that let's someone new to a field quickly gain context for what is being done, what has been done, and what they need to do to develop their knowledge and skills to get to a point where they can contribute. And we want to do it in a way that eliminates unnecessary obstacles like jargon. We cannot promise to teleport a student to being at the forefront of their field, but we can promise to clear the path, and allow them to see where they need to go.
Our project provides a visual representation of published papers in any field, allowing users to see what material they need to cover to understand some specific paper. In addition, we use cutting edge LLM models to generate simplified summaries of article abstracts, allowing us to provide the core information of the paper in a way that caters to any reading level. We also ensure to provide numerous external resources, for further reading in a variety of media formats. At the end of the day, we want to be the first place a novice, aspirational researcher goes to learn a new topic.
Log in or sign up for Devpost to join the conversation.