Continuous Prompt Generation from Linear Combination of Discrete Prompt Embeddings

arXiv:2312.10323v2 Announce Type: replace
Abstract: The wayward quality of continuous prompts stresses the importance of their interpretability as unexpected and unpredictable behaviors appear following training, especially in the context of large language models automating people-sensitive tasks such as resume screening. In this paper we present a novel method of constructing continuous prompts via discrete prompt embeddings and evaluate improvements to continuous prompt interpretability and inference accuracy. For a set of manually designed discrete prompts $\mathcal{D}$, which we tokenize and embed …

abstract arxiv combination context continuous cs.cl embeddings importance interpretability language language models large language large language models linear novel paper people prompt prompts quality resume screening tasks training type