Prompt Optimization via Adversarial In-Context Learning

arXiv:2312.02614v2 Announce Type: replace
Abstract: We propose a new method, Adversarial In-Context Learning (adv-ICL), to optimize prompt for in-context learning (ICL) by employing one LLM as a generator, another as a discriminator, and a third as a prompt modifier. As in traditional adversarial learning, adv-ICL is implemented as a two-player game between the generator and discriminator, where the generator tries to generate realistic enough output to fool the discriminator. In each round, given an input prefixed by task instructions and …

abstract adversarial adversarial learning arxiv context cs.cl cs.lg game generator in-context learning llm optimization prompt type via