LLM Behavior
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
LLM Behavior Subset 1
99.8IR
24
LLM Behavior
95Response Rate (RR)
12
LLM Behavior
100Injection Rate (IR)
10
LLM Behavior Goal-Distant
100IRany
5
LLM Behavior Goal-Adjacent
100IRany
5