Professor Eric Stokan, Director of UMBC's Center for Social Science Scholarship, will lead a session on best practices in prompt engineering using ChatGPT to engage in image detection in R through an API.  He will demonstrate how going from zero-shot prompting with poor prompting language (e.g., lacking schema, rubric, and clear separation of user from system prompts) to multi-shot prompting (using images and human coding) improves the performance.  He wii also describe the prospects of using this for research and some evaluation techniques for estimating the effectiveness of LLMs and LRMs (Large Reasoning Models designed to solve complex problems by "thinking" before they speak).

Lunch will be provided for registered attendees.

REGISTER HERE

Hosted by the Center for Social Science Scholarship and cosponsored by the Division of Information Technology, the Center for Scalable Data and Computational Science, and CGC-SCIPE. CS3-sponsored events are open for full participation by all individuals regardless of race, color, religion, sex, national origin, or any other protected category under applicable federal law, state law, and the University's nondiscrimination policy.