Preprint

TROPT: An Open Framework for Unifying and Advancing Discrete Text Optimization
An open-source framework for running and developing discrete text-trigger optimizers—against any NLP model and any loss. TROPT ships 40+ recipes spanning LLM jailbreaks, model auditing, and interpretability, lowering the barrier to adopting discrete text optimization methods.
Matan Ben-Tov, Mahmood Sharif
   PDF    Code    Project   arXiv
TROPT: An Open Framework for Unifying and Advancing Discrete Text Optimization