Spring 2025 Independent Study

In Spring 2025, the UF Data Studio will host two Independent Study topics.

The independent study topics are an opportunity to investigate a topic further through group-guided research activities. If you are interested, please fill out the appropriate section of the form.

The form submission deadline is Thursday, January 14, at 7 pm ET.

Critical Multimodal Data Sets 

Visual large language models are quickly being deployed and made accessible to those in need. We want to investigate the types of data sets that current state-of-the-art VLMs have difficulty interpreting. The bounds of these visual language models must be explored to avoid critical breakdowns and further that capabilities.

This independent study will develop datasets for VLM experiments.

Evidence of research capability is required, and working knowledge of machine learning is required.

Anomaly Finding Visualization Methods

Anomalies will always appear in data. Statistical anomalies may appear because of errors in entries, surprising events, or emergent trends. Identifying anomalies is important for any downstream use of databases. Researchers will select from a wide range of user interfaces depending on the data and the task (e.g., Gerrymandering)

oIn this Independent Study, we will survey the interfaces and modalities used to identify various anomaly types and research the critical datasets where anomaly discovery can be applied. Evidence of a thorough grasp of machine learning, statistics, and user interfaces is needed.