List your company

Abstention Enforcement on TruthfulQA no-context abstention stress test 100-item

100CorrectnessGPT-4o-mini
-4235077Mar 12, 2026

Evaluation results

MethodLinks
GPT-4o-miniCondition=Instruction2026.031000100
GPT-4oCondition=Instruction2026.031000100
GPT-4oCondition=Composite2026.031000100
GPT-3.5-turboCondition=Gate2026.031000100
GPT-3.5-turboCondition=Composite2026.031000100
GPT-4oCondition=Gate2026.0399199
GPT-4o-miniCondition=Gate2026.0398298
GPT-4o-miniCondition=Composite2026.0398298
GPT-3.5-turboCondition=Instruction2026.03623862
GPT-4o-miniCondition=Baseline2026.0301000
GPT-4oCondition=Baseline2026.0301000
GPT-3.5-turboCondition=Baseline2026.0301000