conftrace_

indirect prompt injection

6 papers

Explore in graph

Also known as

IPI

Co-occurring keywords · papers mentioning both

llm agent (3) large language model (2) security defense (1) safety-utility trade-off (1) tool dependency graph (1) instruction detection (1) defenses (1) masking function (1) trajectory re-execution (1) external tool (1)

Papers

ARGUS: Defending Against Multimodal Indirect Prompt Injection via Steering Instruction-Following Behavior CVPR 2026 Adaptive Attacks Break Defenses Against Indirect Prompt Injection Attacks on LLM Agents NAACL 2025 AGENTVIGIL: Automatic Black-Box Red-teaming for Indirect Prompt Injection against LLM Agents EMNLP 2025 Defending against Indirect Prompt Injection by Instruction Detection EMNLP 2025 IPIGuard: A Novel Tool Dependency Graph-Based Defense Against Indirect Prompt Injection in LLM Agents EMNLP 2025 MELON: Provable Defense Against Indirect Prompt Injection Attacks in AI Agents ICML 2025