Injected triggers
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Injected triggers Llama-3.1 70B
2,462Trigger Count
6
Injected triggers Llama-3.1 8B
3,428Rule Adherence Count
6
Injected triggers GPT-4o
4,090Trigger Count (Rules)
6
Injected triggers GPT-4o mini
2,251Rule Adherence Count
6
Injected triggers GPT-3.5 Turbo
1,429Test Rule Count
6