arXiv CS AI
techCenter
Nemotron 3.5 Content Safety Moderator: A Compact Multimodal, Multilingual, and Reasoning Enabled Content Safety Moderatortranslating…
arXiv:2608.27548v1 Announce Type: new
Abstract: Safety moderation for deployed AI applications is moving beyond text-only prompts: systems increasingly need to judge images, documents, screenshots, and generated responses under policies that vary across domains. Existing…
Keywords#Safety#Content#Content Safety#Content Safety Moderator#Safety Moderator