PL-Guard: Probabilistic Logic Reasoning for LLM Guardrails

Abstract

Large language model guardrails can be viewed as policy-consistency problems: a system must determine which policy-relevant facts hold in a prompt-response pair and what those facts imply under a given policy. Common approaches, including policy prompting and LLM-as-a-judge pipelines, often overlap the tasks of semantic grounding and policy reasoning: the model both interprets the prompt-response pair and reasons about whether a policy has been violated. This can lead to unsafe compliance with h

Categories

@misc{chatterji2026plguard,
  title = {{PL-Guard: Probabilistic Logic Reasoning for LLM Guardrails}},
  author = {Satchit Chatterji and Shi-Han Wang and G. Sileno and Erman Acar},
  year = {2026},
  month = aug,
  eprint = {2608.15673},
  archivePrefix = {arXiv},
  url = {https://www.semanticscholar.org/paper/b112fd1f674c4d19bde5fca334828d52a986db60}
}