Prompt Injection Detection is Regime-Dependent: A Deployment-Aware Evaluation with Interpretable Structural Signals

Abstract

Prompt injection poses a critical threat to the safe deployment of large language models, yet existing detection approaches are typically evaluated under limited settings that do not reflect real-world operating constraints. In this work, we present a deployment-aware evaluation of prompt injection detection using a multi-model and multi-regime experimental framework. We compare lexical, semantic, structural, and transformer-based detectors across multiple out-of-distribution settings, repeated

Categories

Framework mappings

Suggested from the entry's categories.

@misc{akinrele2026prompt,
  title = {{Prompt Injection Detection is Regime-Dependent: A Deployment-Aware Evaluation with Interpretable Structural Signals}},
  author = {Akindoyin Akinrele and Shreyank N Gowda},
  year = {2026},
  month = may,
  eprint = {2605.26999},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2605.26999}
}