Do you want to equip the social sciences for the looming AI revolution in academic literature exploration by investigating the best AI model for ASReview? If so, apply for this PhD position! This position is part of the bigger project “Transparent and Reproducible AI-aided Systematic Reviewing for the Social Sciences (TRASS)”.
Wat ga je doen? The rapidly evolving field of AI offers promising solutions to the literature screening challenge using machine learning models, like active learning, and, very recently, large language models (LLMs). However, many of these AI-driven solutions emerge from tech companies that publish new models at an unprecedented rate. The rapidity of advancements in the field of AI outpaces meticulous scientific evaluations, leaving many methods unrefined and unproven. The challenge is twofold: keeping pace with these relentless innovations while collaboratively forging a comprehensive understanding of their strengths and limitations.
In the evolving landscape of AI-aided systematic tools (like
ASReview), we need to explore which AI model can best be used for which type of data. For instance, while the active learning model (ALM) can facilitate literature screening by presenting the most likely relevant record, the human still needs to make the final labeling decision. In contrast, an LLM can directly generate such labels with explanations, without human input. This project envisions a collaborative approach that leverages the strengths of both humans and different artificial intelligence solutions in systematic screening.
You will be responsible for carrying out several high-quality simulation studies combining ALM/LLMs and testing the performance on 500+ labeled datasets.