BiasKG: Adversarial Knowledge Graphs to Induce Bias in Large Language Models

arXiv:2405.04756v1 Announce Type: cross
Abstract: Modern large language models (LLMs) have a significant amount of world knowledge, which enables strong performance in commonsense reasoning and knowledge-intensive tasks when harnessed properly. The language model can also learn social biases, which has a significant potential for societal harm. There have been many mitigation strategies proposed for LLM safety, but it is unclear how effective they are for eliminating social biases. In this work, we propose a new methodology for attacking language models …

abstract adversarial arxiv bias biases commonsense cs.cl cs.lg graphs harm knowledge knowledge graphs language language model language models large language large language models learn llms modern performance reasoning social tasks type world