Safeguarding AI: Effectiveness of Guardrails in Controlling Malicious Output from Locally Hosted LLMs
This paper explores the effectiveness of open-source guardrails that can be added to LLM-based conversational applications to mitigate the threat of potential misuse.
