Research on how to safely use large language models in a user-facing application without embarrassing your organization.
A safe digital assistant
Research results on building a safe digital assistant using open-source LLMs and parameter-efficient fine-tuning techniques.