AutoRISE: Agent-Driven Strategy Evolution for Red-Teaming Large Language Models

Abstract

Automated red-teaming methods for large language models typically optimize attack prompts within a fixed, human-designed strategy, leaving the attack strategy itself unchanged. We instead optimize the strategy. We propose AutoRISE, a method that searches over executable attack programs rather than individual prompts. At each iteration, a coding agent edits a strategy and a fixed evaluation harness scores the resulting attacks, returning both a scalar objective and per-example diagnostics that gu

Categories

Framework mappings

Suggested from the entry's categories.

@misc{gautam2026autorise,
  title = {{AutoRISE: Agent-Driven Strategy Evolution for Red-Teaming Large Language Models}},
  author = {Tanmay Gautam and Alireza Bahramali and Sandeep Atluri},
  year = {2026},
  month = apr,
  eprint = {2604.22871},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2604.22871}
}