MORL-Prompt: An Empirical Analysis of Multi-Objective Reinforcement Learning for Discrete Prompt Optimization

arXiv:2402.11711v1 Announce Type: new
Abstract: RL-based techniques can be used to search for prompts that when fed into a target language model maximize a set of user-specified reward functions. However, in many target applications, the natural reward functions are in tension with one another -- for example, content preservation vs. style matching in style transfer tasks. Current techniques focus on maximizing the average of reward functions, which does not necessarily lead to prompts that achieve balance across rewards -- an …

abstract analysis applications arxiv cs.cl example fed functions language language model multi-objective natural optimization prompt prompts reinforcement reinforcement learning search set type