ROSE Doesn't Do That: Boosting the Safety of Instruction-Tuned Large Language Models with Reverse Prompt Contrastive Decoding

arXiv:2402.11889v1 Announce Type: new
Abstract: With the development of instruction-tuned large language models (LLMs), improving the safety of LLMs has become more critical. However, the current approaches for aligning the LLMs output with expected safety usually require substantial training efforts, e.g., high-quality safety data and expensive computational resources, which are costly and inefficient. To this end, we present reverse prompt contrastive decoding (ROSE), a simple-yet-effective method to directly boost the safety of existing instruction-tuned LLMs without any additional training. The …

abstract arxiv become boosting cs.cl current decoding development instruction-tuned language language models large language large language models llms prompt quality safety training type