LongPIBench: A Long-Context Benchmark for Prompt Injection
尚未评估本次未安排独立复现局限性claim-scope-limitations
The benchmark evaluates static document-centric workflows in which the full document is supplied in one inference call and does not cover dynamic multi-step agentic workflows or the full range of automated attacks.
来源：paper:p. 12, Limitations
报告指标与观测值
这条 Claim 暂无结构化指标。
Assessment（0）
这条 Claim 暂无已发布的不可变 Assessment。
关联运行（0）
这条 Claim 暂无关联执行。