Evaluating the role of `Constitutions' for learning from AI feedback
This work addresses the problem of optimizing AI feedback for training LLMs in healthcare communication, but it is incremental as it highlights limitations in certain areas.
The study investigated how different constitutions affect AI feedback quality for improving patient-centered communication in medical interviews, finding that detailed constitutions enhanced emotive qualities but none outperformed the baseline in practical skills like information gathering.
The growing capabilities of large language models (LLMs) have led to their use as substitutes for human feedback for training and assessing other LLMs. These methods often rely on `constitutions', written guidelines which a critic model uses to provide feedback and improve generations. We investigate how the choice of constitution affects feedback quality by using four different constitutions to improve patient-centered communication in medical interviews. In pairwise comparisons conducted by 215 human raters, we found that detailed constitutions led to better results regarding emotive qualities. However, none of the constitutions outperformed the baseline in learning more practically-oriented skills related to information gathering and provision. Our findings indicate that while detailed constitutions should be prioritised, there are possible limitations to the effectiveness of AI feedback as a reward signal in certain areas.