Unveiling the Misuse Potential of Base Large Language Models via In-Context Learning

arXiv:2404.10552v1 Announce Type: new
Abstract: The open-sourcing of large language models (LLMs) accelerates application development, innovation, and scientific progress. This includes both base models, which are pre-trained on extensive datasets without alignment, and aligned models, deliberately designed to align with ethical standards and human values. Contrary to the prevalent assumption that the inherent instruction-following limitations of base LLMs serve as a safeguard against misuse, our investigation exposes a critical oversight in this belief. By deploying carefully designed demonstrations, our research …

abstract alignment application arxiv context cs.ai cs.cl datasets development ethical human in-context learning innovation language language models large language large language models llms misuse progress scientific standards type values via