List your company

Cultural norm reasoning on NormAd West Asia 1.0 (test)

69.53AccuracyQwen3.5-4B
52.671657.048361.42565.8017Mar 12, 2026

Evaluation results

MethodLinks
Qwen3.5-4BPrompt Language=Source-Language Stories2026.0369.53
Tiny Aya WaterPrompt Language=Source-Language Stories2026.0369.27
Tiny Aya FirePrompt Language=Source-Language Stories2026.0368.87
SmolLM3-3BPrompt Language=Source-Language Stories2026.0368.83
Tiny Aya GlobalPrompt Language=Source-Language Stories2026.0367.44
Qwen3-4BPrompt Language=Source-Language Stories2026.0366.47
Tiny Aya EarthPrompt Language=Source-Language Stories2026.0366.35
Gemma3-4BPrompt Language=Source-Language Stories2026.0363.11
Ministral-3-3BPrompt Language=Source-Language Stories2026.0353.32