@misc{indiciae347efa25792f,
  title = {Agentic Adversarial Rewriting Exposes Architectural Vulnerabilities in Black-Box NLP Pipelines},
  author = {Mazal Bethany and Kim-Kwang Raymond Choo and Nishant Vishwamitra and Peyman Najafirad},
  year = {2026},
  url = {https://arxiv.org/abs/2604.23483},
  note = {Source identifier: 2604.23483}
}