Skip to content
Search
paperOctober 2023ReviewedOpen access

NeMo Guardrails: A Toolkit for Controllable and Safe LLM Applications with Programmable Rails

Traian Rebedea, Razvan Dinu, Makesh Narsimhan Sreedhar, Christopher Parisien, Jonathan Cohen

EMNLP 2023 Demo

View resourcePDF

Abstract

Presents NeMo Guardrails, an open-source toolkit for adding programmable safety, security, and privacy rails to LLM-based conversational systems.

Categories

GuardrailsInput FilteringOutput Moderation
#NeMo#programmable-rails#toolkit

Framework mappings

OWASP Top 10 for LLM Applications
  • LLM01Prompt Injection
  • LLM05Improper Output Handling
  • LLM06Excessive Agency

Cite

@inproceedings{rebedea2023nemo,
  title = {{NeMo Guardrails: A Toolkit for Controllable and Safe LLM Applications with Programmable Rails}},
  author = {Traian Rebedea and Razvan Dinu and Makesh Narsimhan Sreedhar and Christopher Parisien and Jonathan Cohen},
  year = {2023},
  month = oct,
  booktitle = {EMNLP 2023 Demo},
  eprint = {2310.10501},
  archivePrefix = {arXiv},
  doi = {10.18653/v1/2023.emnlp-demo.40},
  url = {https://arxiv.org/abs/2310.10501}
}