Value Alignment
Our research explores ethical dilemmas through diverse prompt engineering.
Model Evaluation
We evaluate GPT-4 and GPT-3.5 using a customized value alignment index.
Fine-Tuning
Selected models are fine-tuned to optimize alignment with specific value systems.
Aligning Values Together
Exploring ethical dilemmas through innovative research and multidisciplinary collaboration for responsible AI development.