Benjamin Seiler

Benjamin Seiler is a postdoctoral research fellow in the department of Epidemiology and Population Health at the Stanford School of Medicine with Mike Baiocchi. He specializes in developing and deploying interpretable, while still effective, statistical learning methods. As part of the Stanford Human Trafficking Data Lab, Dr. Seiler currently works on quantitative approaches to issues of labor trafficking and child labor in Brazil. He holds a PhD in Statistics from Stanford University where he was advised by Art B. Owen.

King Center Supported Research

2023 - 2024 Academic Year | Global Development Research Funding

Large Language Models for Efficient Human Trafficking Tip Processing

Anti-trafficking law enforcement agencies in Brazil rescue thousands from modern slavery each year. To do so, they must review and prioritize large numbers of tips submitted by workers and civil society advocates. Large language models (LLMs) have the potential to augment and accelerate ongoing work to improve and automate this prioritization process. However, the safe and ethical development and testing of LLMs in this setting require novel human-labeled data and input from front-line stakeholders, especially survivor advocacy groups. This project aims to center stakeholders in the rigorous evaluation of the potential for LLM deployment in this high-stakes setting.