Llama

The 'Llama' dataset/benchmark contains a variety of tasks used to evaluate the effectiveness of injection detection in large language models (LLMs) against domain-camouflaged injection attacks.

Papers using Llama (3)

Blind Spots in the Guard: How Domain-Camouflaged Injection Attacks Evade Detection in Multi-Agent LLM Systems2026Agentic Reinforcement Learning for Search Misaligns Instruction-Tuning2025Bridging On-Device and Cloud LLMs for Collaborative Reasoning: A Unified Methodology for Local Routing and Post-Training2025