Projects

This project investigates the capabilities of large language models (LLMs) in multi-agent environments, focusing on their ability to collaborate, compete, and adapt in complex scenarios. We conduct a comprehensive study using a custom-built multi-agent simulation platform. Our findings reveal insights into the strengths and limitations of LLMs in multi-agent settings, highlighting their potential for real-world applications and identifying areas for future research.

RISE: Interactive Visual Diagnosis of Fairness in Machine Learning Models. An interactive system that converts sorted residuals into interpretable patterns for post-hoc fairness diagnosis under domain shift.

RISE: Interactive Visual Diagnosis of Fairness in Machine Learning Models. An interactive system that converts sorted residuals into interpretable patterns for post-hoc fairness diagnosis under domain shift.

This project centers on machine interpretation of Mixtec codexs left to history by the precolonial and early colonial peoples of mesoamerica. It is a continuation of work for our Mixtec NLP grant.

This project is about Accented Speech and Audio Processing using Audio Language models.

This project is about utilizing vision language models (VLMs) to read comic book pages for blind and visually impaired users.