AEGIS: Online Adaptive AI Content Safety Moderation with Ensemble of LLM Experts

arXiv:2404.05993v1 Announce Type: new
Abstract: As Large Language Models (LLMs) and generative AI become more widespread, the content safety risks associated with their use also increase. We find a notable deficiency in high-quality content safety datasets and benchmarks that comprehensively cover a wide range of critical safety areas. To address this, we define a broad content safety risk taxonomy, comprising 13 critical risk and 9 sparse risk categories. Additionally, we curate AEGISSAFETYDATASET, a new dataset of approximately 26, 000 human-LLM …

abstract adaptive ai arxiv become benchmarks cs.cl cs.cy cs.lg datasets ensemble experts generative language language models large language large language models llm llms moderation quality risks safety type