CoLLEGe: Concept Embedding Generation for Large Language Models

arXiv:2403.15362v1 Announce Type: new
Abstract: Current language models are unable to quickly learn new concepts on the fly, often requiring a more involved finetuning process to learn robustly. Prompting in-context is not robust to context distractions, and often fails to confer much information about the new concepts. Classic methods for few-shot word learning in NLP, relying on global word vectors, are less applicable to large language models. In this paper, we introduce a novel approach named CoLLEGe (Concept Learning with …

abstract arxiv college concept concepts context cs.ai cs.cl current distractions embedding few-shot finetuning fly information language language models large language large language models learn process prompting robust type word