LLMGuard: Guarding Against Unsafe LLM Behavior

Shubh Goyal; Medha Hira; Shubham Mishra; Sukriti Goyal; Arnav Goel; Niharika Dadu; Kirushikesh DB; Sameep Mehta; Nishtha Madaan

LLMGuard: Guarding Against Unsafe LLM Behavior

Computation and Language 2024-03-05 v1 Cryptography and Security Machine Learning

Authors: Shubh Goyal , Medha Hira , Shubham Mishra , Sukriti Goyal , Arnav Goel , Niharika Dadu , Kirushikesh DB , Sameep Mehta , Nishtha Madaan

View on arXiv ↗ PDF ↗

Abstract

Although the rise of Large Language Models (LLMs) in enterprise settings brings new opportunities and capabilities, it also brings challenges, such as the risk of generating inappropriate, biased, or misleading content that violates regulations and can have legal concerns. To alleviate this, we present "LLMGuard", a tool that monitors user interactions with an LLM application and flags content against specific behaviours or conversation topics. To do this robustly, LLMGuard employs an ensemble of detectors.

Keywords

ai safety large language model vulnerability detection

Cite

@article{arxiv.2403.00826,
  title  = {LLMGuard: Guarding Against Unsafe LLM Behavior},
  author = {Shubh Goyal and Medha Hira and Shubham Mishra and Sukriti Goyal and Arnav Goel and Niharika Dadu and Kirushikesh DB and Sameep Mehta and Nishtha Madaan},
  journal= {arXiv preprint arXiv:2403.00826},
  year   = {2024}
}

Comments

accepted in demonstration track of AAAI-24

LLMGuard: Guarding Against Unsafe LLM Behavior

Abstract

Keywords

Cite

Comments

Related papers