Jade Franklin

Jade at grand canyon

Jade is a computer science PhD student under Deborah McGuinness. She's been at Rensselaer since 2016, having graduated in 2019 with her bachelor's in CS, and continued at RPI in 2020 to pursue research in knowledge graph applications.

She has worked on research projects in a variety of domains, including biomedical knowledge extraction, fairness in machine learning, explainable AI, and fact-checking of LLMs. Her primary research interest is in the use of ontologies to guide AI in procedural content generation, and she is especially excited by the possibilities presented by neurosymbolic foundation models to enhance media creation workflows. However, she is increasingly concerned about the dangers of LLMs replacing human-created content with subpar, inaccurate, or biased content, as well as the dangers that LLMs pose to artistic livelihoods. As an artist and writer herself, she understands the existential threat posed to these careers, and seeks to understand through her research how current negative trends in the AI field can be reversed.

Her hobbies include spending far too much time thinking about hexagons and resisting the urge to complain about minor grammar errors.

Affiliations: Tetherless World Constellation

Projects as Research Assistant

ChatBS: A Context-aware LLM Exploratory Sandbox uses the OpenAI Completion API service (GPT-4-0613 model) to answer questions.

This is a program led by the Defense Advanced Research Projects Agency (DARPA), which is developing a schema-based AI system that can identify complex events and bring them to the attention of users.

The Center for Health Empowerment by Analytics, Learning, and Semantics (HEALS) is a five-year collaboration between Rensselaer and IBM aimed at researching how the application of advanced cognitive computing capabilities can help people to understand and improve their own health conditions.