List your company

Over-refusal Assessment on XS (test)

93.1F1 ScoreSelf-Reminder
50.14861.29972.4583.601Jan 31, 2026

Evaluation results

MethodLinks
Self-ReminderModel Scale=Qwen3-14B2026.0193.1
Self-ReminderModel Scale=Qwen3-8B2026.0192.6
Self-ReminderModel Scale=Qwen3-4B2026.0192.5
ReasoningGuardModel Scale=Qwen3-4B2026.0192.4
Self-GuardModel Scale=Qwen3-8B2026.0192
ReasoningGuardModel Scale=Qwen3-8B2026.0191.3
SafeKeyModel Scale=Qwen3-14B2026.0191
Self-GuardModel Scale=Qwen3-14B2026.0191
STAR-1Model Scale=Qwen3-14B2026.0189.6
Self-GuardModel Scale=Qwen3-4B2026.0189.3
SafeKeyModel Scale=Qwen3-8B2026.0189.3
STAR-1Model Scale=Qwen3-8B2026.0188.4
Alpha-steerModel Scale=Qwen3-14B2026.0187.4
InstructModel Scale=Qwen3-14B2026.0187.1
ReasoningGuardModel Scale=Qwen3-14B2026.0186.7
Alpha-steerModel Scale=Qwen3-4B2026.0186.3
InstructModel Scale=Qwen3-4B2026.0186.1
STAR-1Model Scale=Qwen3-4B2026.0185.2
Alpha-steerModel Scale=Qwen3-8B2026.0184.5
SafeKeyModel Scale=Qwen3-4B2026.0184.1
InstructModel Scale=Qwen3-8B2026.0183.1
SafeChainModel Scale=Qwen3-14B2026.0169.1
SafeChainModel Scale=Qwen3-8B2026.0163.5
SafeChainModel Scale=Qwen3-4B2026.0151.8