No AI summary available for this article.
Why It Matters
Prompt injection attacks pose a serious security risk to large language models in real-world applications.
Provenance
Discovered via ArXiv and published by ArXiv.
Key Claims
Original description
Prompt injection attacks pose a serious security risk to large language models in real-world applications. However, existing prompt injection benchmarks primarily focus on short-context inputs, leaving the attacks and defenses in long-context settings largely unexplored. This gap leads to a substantial overestimation of the effectiveness of current defenses. In this paper, we bridge the gap by introducing LongPIBench, a long-context benchmark for prompt injection covering 4 realistic application scenarios: paper peer review, resume screening, code review, and email summary. For each scenario,...
Discovered via ArXiv
Research papers and preprints from arXiv.
Publisher: arxiv.org
ID: http://arxiv.org/abs/2608.28411v1 · Indexed 13 days ago