conftrace_

word importance ranking

2 papers

Explore in graph

Also known as

WIR

Co-occurring keywords · papers mentioning both

black-box adversarial attack (1) adversarial text attack (1) black-box scenario (1) neural network vulnerability (1) large language model (1) guardrail system (1) character injection (1) jailbreak detection (1) word substitution (1) prompt injection (1)

Papers

Bypassing LLM Guardrails: An Empirical Analysis of Evasion Attacks against Prompt Injection and Jailbreak Detection Systems ACL 2025 PARSE: An Efficient Search Method for Black-box Adversarial Text Attacks COLING 2022