Sharpe Ratio-Guided Active Learning for Preference Optimization in RLHF

Re-assign community