FlashRT: Towards Computationally and Memory Efficient Red-Teaming for Prompt Injection and Knowledge Corruption

Abstract

Long-context large language models (LLMs)-for example, Gemini-3.1-Pro and Qwen-3.5-are widely used to empower many real-world applications, such as retrieval-augmented generation, autonomous agents, and AI assistants. However, security remains a major concern for their widespread deployment, with threats such as prompt injection and knowledge corruption. To quantify the security risks faced by LLMs under these threats, the research community has developed heuristic-based and optimization-based r

Categories

Framework mappings

Suggested from the entry's categories.

@misc{wang2026flashrt,
  title = {{FlashRT: Towards Computationally and Memory Efficient Red-Teaming for Prompt Injection and Knowledge Corruption}},
  author = {Yanting Wang and Chenlong Yin and Ying Chen and Jinyuan Jia},
  year = {2026},
  month = apr,
  eprint = {2604.28157},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2604.28157}
}